Quantum Computing for Biomedical Computational and Data Sciences: A

Joint DOE-NIH Roundtable Report

Report on the 2-day roundtable held on March 17 and March 27, 2023

Co-chairs: Dr. Shannon McWeeney, Dr. Talita Perciano

DOE (OS/ASCR): Dr. Ceren Susut, Dr. Lali Chatterjee, Dr. Marco Fornari
NIH (OD/DPCPSI/ODSS): Dr. Laura Biven, Dr. Christopher Siwy

https://doi.org/10.2172/2228574

December 20, 2023

Contents

1 Executive Summary 2
1.1 Background and Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Key Outcomes and Research Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2.1 Biomedical Use-Cases for Quantum Applicability and Impact . . . . . . . . . . . . . . . . . . . . 2
1.2.2 Cross-Cutting Supporting Research Topics - Current State of the Field and What is Needed . . . . . 3
1.2.3 Examples of Promising Research Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.4 List of Roundtable Participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

2 Introduction 6
2.1 The Role of Quantum Computing in Biomedical Computational and Data Sciences . . . . . . . . . . . . . 6
2.2 DOE and NIH Collaboration: Aims and Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3 Roundtable Participants and Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3.1 Participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

3 Roundtable Perspective on Current State of Quantum Computing and Biomedical Research 9

3.1 Discussed Quantum Computing Technologies and Advancements . . . . . . . . . . . . . . . . . . . . . . 9
3.2 Discussed Existing Quantum Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.3 What are the applications that are amenable for the shift from Classical to Quantum? . . . . . . . . . . . . 10
3.4 Existing Collaborations and Projects in the Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

4 Biomedical Use-Cases for Short-Term Quantum Applicability and Impact 12

4.1 Medical Imaging and Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.1.1 Basic Image Analysis Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
4.1.2 Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.2 Biological Sequence Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.3 Drug Discovery and Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

5 Discussed Cross-Cutting Supporting Research Topics 15

5.1 Data Preprocessing, Encoding/Decoding, and Input/Output Data . . . . . . . . . . . . . . . . . . . . . . . 15
5.2 Quantum Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
5.3 Machine Learning and Neural Networks in Biomedical Research . . . . . . . . . . . . . . . . . . . . . . . 16
5.4 Foundational Mathematical Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5.5 Addressing Noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5.6 Privacy Preserving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
5.7 Quantum Simulation and Real Quantum Hardware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
5.8 Workforce Development and Accessibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
5.9 Ethics and Governance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

6 Conclusion 23

7 Addendum 24

1
1. Executive Summary
1.1 Background and Objectives
The overlap of quantum computing and biomedical research, while less explored, presents significant near-term opportu-
nities. The Department of Energy (DOE) and the National Institutes of Health (NIH) are interested in exploiting the DOE
community’s capabilities and expertise in quantum computing to potentially advance biomedical research, targeting funda-
mental studies of biological and molecular structures, understanding of human health as well as mental and physical disorders
and diseases, and deriving insights from clinical data. NIH’s approach to quantum computing is guided by its Strategic Plan
for Data Science, emphasizing the importance of findable, accessible, interoperable, and reusable (FAIR) data assets, security
and privacy of data, and efficient computing and storage. DOE’s Office of Science (SC), and more specifically the Advanced
Scientific Computing Research (ASCR) program, supports quantum information science (QIS) research, contributing to a
unique portfolio of quantum computing and communications expertise.
This roundtable was assembled to consider the opportunities and challenges in the near-, medium-, and long-term at
the intersection of quantum computing, data science, and biomedical research and how these could be addressed through
inter-agency collaboration and multi-disciplinary partnerships.
1.2 Key Outcomes and Research Directions
Experts in quantum computing and biomedical research gathered in a roundtable to highlight significant opportunities and
challenges in these intersecting domains. Presentations and discussions led to the identification of potential biomedical ap-
plications of quantum computing, assessing current research status, and considering technical and practical aspects. This
section summarizes the promising research directions and strategies to tackle identified challenges, guiding future collabo-
rations between DOE and NIH to foster groundbreaking quantum solutions for biomedical research.
1.2.1 Biomedical Use-Cases for Quantum Applicability and Impact
The roundtable discussions identified potential biomedical use-cases for quantum computing in the near-term, i.e., although
many of the quantum challenges are medium- and long-term, there could be initial applicability and impact in the next 5
years (near-term). Notably, three fields stood out due to their preliminary work and anticipated impact: medical imaging
and diagnostics, biological sequence analysis and processing, and drug discovery. There was recognition of the complexity
and diversity within these fields, including the variety of algorithms employed depending on the data type and research
question. Crucially, steps where classical computation restricts problem size [1], and could hence benefit from a quantum
advantage [2], must be highlighted.
Medical Imaging and Diagnostics. This field could greatly benefit from quantum algorithms as traditional methods for
image analysis (e.g., segmentation), reconstruction (e.g., iterative methods), and registration (e.g, feature/intensity/distance-
based methods) are computationally demanding, and the need for improved diagnostic accuracy is paramount. Quantum al-
gorithms, including quantum Fourier transform [3], quantum phase estimation, quantum optimization methods [4], quantum
machine learning (QML) [5], and quantum image processing techniques [6], can be adapted to enhance medical imaging
applications. The demand for quicker, more accurate, and efficient diagnostics is high, and specific applications, such as
image denoising and filtering applied to smaller scale classical and quantum input, may be feasible with near-term quan-
tum hardware due to their modest qubit requirements, especially the initial study and development of proof-of-principle
demonstrations.
Biological Sequence Data Analysis. The process of extracting meaningful biological insights from raw DNA/RNA se-
quencing data involves tasks such as identifying genes, detecting mutations, and understanding gene expression patterns.
There is a great need for faster sequence alignment and search, enhanced phylogenetic analysis, and multi-omics data inte-
gration. This field has potential for short-term impact by the use of quantum algorithms, including the ones based on distance
metrics such as Edit distance [7] and Hamming distance [8], and quantum random walks [9], which could be directly applied
to sequence search and sequence matching. Moreover, there is an extremely high-demand for efficient analysis in this field,
making the need for more efficient algorithms urgent.
Drug Discovery and Design. Computer-aided drug design (CADD) is used in the drug discovery process to identify and
develop a potential lead. This process is computationally challenging given the combinatorial nature of the problem and
the extensive simulation timescales that may be necessary. Methods based on quantum simulation and optimization [10],
QML [11, 12], and hybrid quantum-classical algorithms [13] are the most promising near-term approaches in this area. For
example, the work proposed in [13] used a hybrid quantum-classical workflow combining classical docking and molecular
dynamics with quantum machine learning to find ligands binding to proteins.

2
1.2.2 Cross-Cutting Supporting Research Topics - Current State of the Field and What is Needed
This section presents the roundtable’s view on advancing quantum computing for biomedical applications summarizing the
most crucial and relevant cross-cutting topics supporting this research endeavor.
Data Preprocessing, Encoding/Decoding, and Input/Output Data. Quantum data encoding methods to represent classical
data in quantum states exist [14, 15, 16, 17, 18, 8] but their use in biomedical sciences is in early exploration stages.
Data preprocessing and encoding for biomedical sciences require efficient and robust techniques to translate complex, high-
dimensional data into quantum states while preserving relevant information. Advances needed include efficient encoding
schemes that reduce qubit usage, gate complexity, and circuit depth. Additional considerations include the use of not only
classical input but also quantum input, which would be quantum data directly obtained from quantum sensors. Despite
the “Big Data” aspect, some problems involve small data sets, posing unique challenges like limited statistical power, data
quality, and a low signal to noise ratio, which often leads to greater uncertainty.
Quantum Algorithms. Quantum optimization algorithms have the potential for solving problems more efficiently than their
classical counterparts. These include, e.g., Quantum Approximate Optimization Algorithms (QAOA) [19], optimization al-
gorithms based on Grover’s search [20], quantum data fitting, quantum semidefinite programming, quantum annealing [20],
and adiabatic quantum computing [21]. Initial research could focus on tailoring existing algorithms to address specific
biomedical challenges, developing noise-resilient algorithms, and exploring hybrid approaches. QML leverages the unique
capabilities of quantum computing to develop and enhance machine learning algorithms to provide speedups and novel ap-
proaches to data analysis problems. There are already well-established QML algorithms such as quantum support vector ma-
chine (QSVM) [22], variational quantum classifier (QVC) [23], quantum (convolutional) neural networks (Q(C)NNs) [24],
and quantum transfer learning [25]. Also, some existing applications to biomedical sciences have been explored for breast
cancer detection [26, 27], molecular docking process [28], and medical imaging [29]. However, much more progress is
needed to unveil the potential of QML for biomedical applications. Potential initial steps include adapting QML algorithms
to accommodate specific characteristics of biomedical data, novel ways for data preprocessing and feature extraction, feature
learning, and geometric quantum machine learning [30]. Finally, mathematical foundations are essential to the development
of quantum algorithms, e.g., efficient quantum techniques for performing linear algebraic operations, such as matrix inver-
sion, eigendecomposition, and singular value decomposition. Block encodings [31, 32], for example, serve as a crucial tool
for representing and manipulating matrices in quantum algorithms. These types of techniques form the basis for many quan-
tum algorithms, including QML and optimization. The applicability of these techniques to biomedical sciences lies in the
potential to enable more efficient processing of large-scale data, often encountered in genomics, medical imaging, and drug
discovery. Potential research directions could include adapting quantum-numerical linear algebra techniques [33, 34, 35]
and block encodings to accommodate the specific characteristics, structure, and volume of biomedical data, for example.
Addressing noise. Quantum computing and quantum algorithms can be affected by various sources of errors and inaccu-
racies which can corrupt quantum states and operations. Such errors influence the reliability and performance of quantum
computations, which are crucial for data-sensitive applications. Consequently, a specific area of research in QIS focuses on
developing error mitigation techniques and noise-resilient quantum algorithms to tackle the impact of noise. Particularly for
near-term applications, dealing with noise associated with noisy intermediate-scale quantum (NISQ) devices is a pressing
need. The development of specific error mitigation techniques, noise-robust algorithms, and noise learning techniques are
examples of potential first steps in this area. Also, studying the relationship between inductive bias and quantum error may
play a key role, i.e., the choice of inductive bias can impact the resilience of the quantum algorithm to noise and error,
potentially affecting the algorithm’s performance and its ability to generalize well.
Privacy preserving. Given challenges in data sharing and computation on protected or proprietary data, there is tremendous
interest in privacy-preserving quantum machine learning. Proof-of-principle demonstrations were discussed which could
ensure confidentiality and accurate learning on NISQ technology. These involve hybrid quantum-classical models trained to
preserve privacy using differentially private optimization algorithms that can curtail information leakage from training data,
with improved model accuracy and faster convergence [36]. These efforts are key for providing the groundwork for secure
QML infrastructure and support distributed QML training which can better utilize available NISQ devices [37]. More recent
work has combined quantum Federated Learning (QFL) and Quantum Differential Privacy (QDP) to achieve comprehensive
protection against both data leakage (QFL) and model inversion attacks (QDP) [38] .
Quantum hardware, simulators, software packages. Existing NISQ devices have a limited amount of qubits and high
error rates, restricting their ability to perform large-scale computations. Alongside NISQ devices, quantum simulation tools
have been developed to design, test, and optimize quantum algorithms on classical computers, offering insights into quantum

3
computing capabilities and potential applications. There is an enormous potential for the combined use of NISQ devices,
simulators executed in supercomputers, and software packages to make impactful progress in quantum for biomedical data
processing. However, specific advancements are needed including expanding the range of available quantum software pack-
ages focused on this scientific field and the development of benchmarks. Given the rapid advancement in quantum hardware
technology, initial proof-of-concept algorithms have the potential to scale rapidly to solve large-scale problems in the future.
Workforce development, accessibility, ethics, and governance. Much of the focus to date in quantum-related training and
workforce development has been on QIS, such as DOE efforts and others. However, there are distinct skills and training
considerations for applied quantum biological data science. These include interdisciplinary training, emphasis on team
science and critical evaluation of application of quantum approaches in the context of the biological domain. Access will
also be critical to facilitate uptake of these approaches and application to relevant data science problems both with respect
to access to technology (e.g., cloud, simulators) and cognitive accessibility (e.g., user-friendly software and toolkits that are
integrated into domain specific workflows). The Roundtable also discussed the disruptive potential of quantum if existing
information technology, data, and governance capabilities are not expanded upon. In fact, ethics and governance issues
around Artificial Intelligence (AI) will be naturally carried out to quantum-based algorithms.
1.2.3 Examples of Promising Research Areas
The topics and themes presented above and discussed during the Roundtable may result in several key research directions
related to quantum for biomedical data analysis. The table below presents a few illustrative examples of research areas
that aim to inspire new collaborations to advance the state of the field and foster a deeper understanding of the potential
applications and limitations of quantum computing in biomedical research.

Category Research Area Description

Medical Imaging and Diagnostics Quantum-enhanced medical Developing novel quantum algorithms
imaging analysis to improve denoising, segmentation, and
registration of MRI and CT scans
Biological Sequence Data Analysis Quantum algorithms for genome Designing quantum algorithms to
analysis accelerate genomic data analysis at scale,
allowing population level characterization
and efficient meta-analysis
Drug Discovery and Design Quantum simulation for drug Leveraging quantum devices and simulators
discovery to perform quantum molecular simulations
for drug discovery and protein folding
prediction
Data Preprocessing, Encoding, Quantum data encoding for Investigating efficient quantum data enco-
and Input Data biomedical applications dings and preprocessing techniques for
complex biomedical data types
Quantum Machine Learning Quantum machine learning for Developing quantum machine learning
precision medicine algorithms to guide therapeutic selection
and patient stratification for precision medicine
and diagnostics
Privacy Preserving Privacy-preserving quantum Developing privacy-preserving quantum
algorithms algorithms to protect sensitive biomedical
data during processing and analysis
Accessibility & Workforce Quantum education for biomedical Developing educational resources, training
Development scientists programs, and multidisciplinary collabora-
tions to build a quantum-ready workforce in
in the biomedical field and develop accessible,
integrated quantum workflows

4
1.2.4 List of Roundtable Participants
Roel Van Beeumen (LBNL) Mark Gerstein (Yale) Antonio G. Peña (UCSD) Co-Chairs:
E. Wes Bethel (SFSU) Kathleen Hamilton (ORNL) Carlo Pierpaoli (NIH) Shannon McWeeney (OHSU)
Marco Cerezo (LANL) Mustafa Irfanoglu (NIH) Gregory Quiroz (JHU) Talita Perciano (LBNL)
Benjamin Cordier (OHSU) Jeffrey Larson (ANL) Gaoyuan Wang (Yale)
Fida Dankar (NYU) Carlos O. Marrero (PNNL) Shinjae Yoo (BNL)
Prashant Emani (Yale) Murphy Niu (University of
Maryland)

5
2. Introduction
2.1 The Role of Quantum Computing in Biomedical Computational and Data Sci-
ences
Biomedical data science integrates principles and techniques from computer science, statistics, mathematics, biology
and medicine to extract meaningful insights and knowledge in order to advance our understanding of diseases, devel-
oping new treatments, and improving healthcare delivery. It facilitates evidence-based medicine, precision medicine,
and the integration of data-driven approaches into healthcare systems, ultimately leading to better patient care and
outcomes. Computing is an indispensable component of biomedical data science needed to process, analyze, and
interpret increasingly larger and more heterogenous datasets, enabling data-driven insights and accelerating research.
However, classical computing is facing increasing challenges with regard to scale and complexity. Quantum comput-
ing provides a cutting-edge computational paradigm that leverages the principles of quantum mechanics and could
hold significant promise for biomedical data science by addressing these issues.
2.2 DOE and NIH Collaboration: Aims and Scope
Despite the recent advances in quantum computing and its applications in many scientific fields, the intersection of
quantum computing, and biomedical research remains a less explored research area. The development of quantum
competencies, experience with utilizing prototype quantum computing systems, and the identification of relevant
quantum applications and algorithms in the biomedical sciences are realistic near-term goals that will better position
and benefit the biomedical research community in quantum computing capabilities. Both NIH and DOE are interested
in exploiting the DOE community’s capabilities and expertise in quantum computing to advance biomedical research
and support the NIH objectives, which include, for example, fundamental studies of biological and molecular struc-
tures; improving understanding of human health as well as mental and physical disorders and diseases; and deriving
insights from clinical data and healthcare. In addition to the NIH’s mission in biomedical and behavioral research,
NIH’s interests in quantum computing are also guided by the NIH Strategic Plan for Data Science with strategic priori-
ties in, for example, metadata for FAIR data reuse and data interoperability; privacy and integrity for controlled access
data; and computational and storage efficiency. This roundtable was assembled to consider the opportunities and chal-
lenges in the near-, medium-, and long-term at the intersection of quantum computing, data science, and biomedical
research and how these could be addressed through multi-disciplinary partnerships. DOE’s Office of Science supports
a diverse portfolio of quantum information science (QIS) research, and the DOE community has built up a unique
suite of capabilities and expertise in this area. Office of Science’s Advanced Scientific Computing Research (ASCR)
program invests in a QIS portfolio focused on quantum computing and communications research including quantum
algorithms, mathematical methods, and quantum computer science
2.3 Roundtable Participants and Structure
Two co-chairs were selected by NIH (ODSS) and DOE (ASCR) respectively and worked closely with them to recruit
roundtable participants and develop the roundtable agenda. All participants were provide pre-work with regard to key
questions to consider.

• What are some of the most pressing challenges in biomedical data science computationally?

• What are the most promising application areas that we should focus on initially?

• What are the key opportunities and challenges on the development of quantum algorithms associated with the
identified application areas?

• What would be the main priority basic research topics in quantum algorithms?

• What would be needed for a successful NIH-DOE collaboration to have real engagement on both sides?

• Identify in terms of biomedical use-cases or classes of applications that can only be addressed via interdisci-
plinary teams of biomedical and QIS experts

• What biomedical communities/groups are best brought together with quantum computing experts?

6
 • What potential benefits and research goals may be expected for the biomedical and computational science
communities?

• What part of a biomedical researcher’s workflow could quantum computing best facilitate?

A pre-roundtable discussion was then held to review scope, provide background, and focus on identification of
the relevant biomedical applications/use-cases (including their potential “parallelism” and “quantum-ism”) as well as
current bottlenecks in data analysis workflow that could benefit from quantum (steps, operations, etc). The formal
roundtable discussion comprised of short talks and discussions based on these use cases and pre-work.
2.3.1 Participants
Seventeen panelists along with the two co-chairs contributed to this report. Thy are:
Roel Van Beeumen PhD is a research scientist in the Computational Research Division at Lawrence Berkeley Na-
tional Laboratory Lab. His research interests range from numerical linear algebra and numerical software to quantum
computing and quantum algorithms.
E. Wes Bethel PhD is an Associate Professor in the Computer Science Department at San Francisco State Uni-
versity, and a Research Affiliate at Lawrence Berkeley National Laboratory. His technical interests include high per-
formance scientific computing, scientific visualization and computer graphics, image analysis and computer vision,
scientific machine learning, quantum computing, scientific data science, and their collective application to challenging
scientific data understanding problems. ACM Distinguished Scientist
Marco Cerezo PhD is a staff scientist at Los Alamos National Laboratory. HIs research interests include Quantum
Machine Learning, Quantum Information, Quantum Computing and Quantum Algorithms, for near-term quantum
devices and for simulation o quantum systems.
Benjamin Cordier, PhD is a senior computational biologist at the OHSU Knight Cancer Institute. HIs research
interests include quantum machine learning applications across biology and medicine, cancer genomics, novel ap-
proaches to democratize access to data and accelerate the creation of FAIR (Findable, Accessible, Interoperable, and
Reusable) and ethically sourced data sets for AI/ML.
Fida Dankar PhD is a visiting Associate Professor of Computer Science at NYUAD. Her general research interests
include Machine learning and data privacy. Currently, her research focuses on the development of multidisciplinary
approaches for the private and secure mining of biomedical data (distributed and central) that draws upon methods
from cryptography, statistics, machine learning, federated learning, and synthetic data generation.
Prashant Emani PhD is Associate Research Scientist at Yale University; His reach interests include human health
genomics, bioinformatics and machine learning.
Mark Gerstein PhD is the Albert L Williams Professor of Biomedical Informatics and Professor of Molecular
Biophysics & Biochemistry, of Computer Science, and of Statistics & Data Science at Yale University. His research is
focused on biomedical data science, and he is particularly interested in machine learning, macromolecular simulation,
human genome annotation & disease genomics, and genomic privacy.
Kathleen Hamilton PhD is a research staff member in the Quantum Computational Science Group at Oak Ridge
National Laboratory. Her research is focused on developing algorithms for next-generation processors including:
neuromorphic computing and adiabatic quantum annealing.
Mustafa Irfanoglu PhD is a staff scientist at the National Institutes of Health. His research interests are in the
application of image processing, computer vision and machine learning techniques to medical imaging data, with
focus on MRI.
Jeffrey Larson PhD is a computational mathematician at Argonne National Laboratory. His research centers
on optimization algorithms and their application to challenging problems ranging over quantum computing/sensing,
particle acclerator design/control, and vehicle platooning/routing.
Carlos Ortiz Marrero PhD is a data scientist at the U.S. Department of Energy’s Pacific Northwest National
Laboratory (PNNL) As the PI for the Computing Optimal States via Quantum Neural Networks project, he is working
on problems related to Quantum Computing and Machine Learning for the Mathematics for Artificial Reasoning in
Science (MARS) initiative. His project is focusing on the study and development of quantum algorithms for near-
term hardware. The goal is to showcase how quantum neural networks provide an interpretable search space for
domain-aware learning problems.

7
 Shannon McWeeney PhD (co-chair) is a professor and vice chair and the Chief Data Officer for the OHSU Knight
Cancer Institute. Her research is the intersection of computer science, biostatistics and genetics to develop compu-
tational approaches to solve research bottlenecks and novel ways to visualize and interpret information in order to
identify clinically relevant signatures for patient stratification and treatment.
Murphy Niu PhD is an Assistant Professor of Computer science at the joint Center for Quantum Information
and Computer Science (QuICS) at the University of Maryland. Her research is focused on intelligent quantum con-
trol optimization and metrology, quantum machine learning, quantum algorithm design and near-term quantum error
correction.
Antonio G. Peña PhD is a computational biologist at UC San Diego. His research is focused on approaches to
facilitate the meta-analysis of the largest microbiome studies to date.
Talita Perciano PhD (Co-chair) is a Research Scientist at Lawrence Berkeley National Laboratory. She conducts
research in the areas of quantum algorithms, quantum image processing and machine learning, scientific machine
learning, image analysis and high-performance computing motivated by the incredible challenges around scientific
data generated by computational models, simulations, and experiments.
Carlo Pierpaoli MD, PhD is a Senior Investigator in the Laboratory on Quantitative Medical Imaging at NIH. His
rsearch is aimed at extracting accurate and reproducible biomarkers from data acquired with non-invasive imaging
techniques, primarily Magnetic Resonance Imaging.
Gregory Quiroz PhD is a staff scientist at the Johns Hopkins University Applied Physics Laboratory. His current
research interests include quantum characterization and control, applications of quantum control to quantum algorithm
design, and quantum sensing.
Gaoyuan Wang PhD is a bioinformaticist at Yale University. Her current research interests lie in understanding bi-
ological systems via large-scale datasets using interdisciplinary approaches. She primarily works on the development
and application of tools that provide insights on biological data. She is also interested in the ethical implication of the
rapidly growing field of health technology, in particular issues related to privacy.
Shinjae Yoo PhD is a computational scientist in the Computer Science and Math of Computational Science Initia-
tive at Brookhaven National Laboratory. His research interests include quantum machine learning, scalable scientific
machine learning, and foundation models.

8
3. Roundtable Perspective on Current State of Quantum Computing
and Biomedical Research

3.1 Discussed Quantum Computing Technologies and Advancements

The theoretical promise of quantum computers is highly enticing, as they hold the potential to revolutionize computing
by exploiting the laws of quantum mechanics. Unlike classical computers, quantum computers can use superposition,
entanglement, and interference to perform certain computations exponentially faster. This is the reason why quantum
computers are the only known form of computation expected to violate the strong Church-Turing thesis, which states
that all computing devices are equivalent up to polynomial-factor overheads [39].
The violation of the strong Church-Turing thesis directly implies the possibility of quantum advantages. This
has led to significant interest in the application of quantum computing methods to numerous problems in scientific
computing. For instance, quantum algorithms have been developed to increase the computational efficiency of solving
linear systems of equations, differential equations, optimization problems, and decomposing matrices [40, 41, 42, 43,
44, 45]. If quantum advantages can be established in practice, these methods have the potential to unlock new insights
and possibilities that are not available through classical computing.
However, the full realization of quantum advantages in these methods remains an open challenge, as the develop-
ment of practical quantum hardware is still in its early stages. Nevertheless, these routines underlie many problems
across numerous applied science domains including biomedical research. Of course, to be most beneficial, such ap-
proaches must be tailored to integrate domain context and account for the specific issues arising when dealing with
biomedical applications.
Existing NISQ hardware platforms are limited by two key constraints. First among them is coherence time, which
measures the time duration that a device is able to maintain a coherent quantum state. In practice, this constraint
is largely captured by device error rates, which can be decomposed into (input) state preparation and measurement
(SPAM) error, gate error (both phase and bit-flip errors), and emergent device error (such as cross-talk between qubits
and interactions between the qubits and their environment). Together, these errors limit the number of gates that can
be executed by a quantum algorithm. The second constraint relates to the number of qubits (currently on the order of
200 to 400) and the connectivity map between them. These two characteristics of a NISQ device directly imply both
i) representational constraints on the size of soluble problem instances and ii) qubit routing overheads [46] (e.g. when
a quantum algorithm assumes all-to-all connectivity, practical implementations may require additional quantum gates
or qubit shuttling to emulate full connectivity), respectively.
Recent advancements in quantum software and hardware have allowed for the development of methods to mitigate
these key constraints. For example, methods to improve the signal-to-noise ratio of quantum circuit outputs, such as
error mitigation [47, 48, 49] and postselection [50, 51], now being used to implement computations of increasing
complexity. Similarly, methods to deconstruct the components of quantum simulation, optimization, and machine
learning problems and map them to quantum circuits that can be executed in parallel [52, 53, 54, 55, 56] offer a path
towards quantum computations beyond the scale of individual NISQ devices while retaining much of the potential
for practical advantages. These examples represent a small sample of the many methods in development to expand
the real-world capabilities of NISQ devices. Finally, while fully error corrected devices are still not available, early-
stage demonstration of error protected qubits are becoming more common (e.g. see [57, 58, 59]). These observations
also point at the possibility of utilizing hybrid devices composed of some error corrected qubit coupled to non-error
corrected ones.
Given the limitations of current quantum hardware, it is important to think about how these would impact the
performance of algorithms aimed at biomedical applications. Specifically, embedding high-resolution images into
quantum states will likely require a large number of qubits, and perhaps prohibitively deep circuits. Developing
algorithms aware of the limitations imposed by near-term devices will be crucial.
3.2 Discussed Existing Quantum Algorithms
A wide variety of quantum algorithms have been developed targeting a range of applications and scopes. Many
foundational algorithms, such as Grover’s algorithm [20], the quantum Fourier transform [60], and the quantum
walk [61] realize quantum computational primitives; in these cases, unstructured search, mapping from the time

9
to frequency domain, and graph sampling and property estimation, respectively. Similarly, generalizations of these
algorithms have expanded the set of primitives that can be realized in a quantum computation. For example, amplitude
amplification and estimation [62] provides a means of implementing more general quantum search algorithms and
quantum counting. By implementing computational primitives, these quantum algorithms provide key operations
that are relevant to larger quantum algorithms and subroutines. For example, Grover’s algorithm is leveraged by
many quantum algorithms targeting NP-hard optimization problems; the quantum Fourier transform and its inverse
are exploited for matrix inversion (a key subroutine in many machine learning algorithms) and phase estimation for
quantum simulation; quantum walks provide a means for computing graph properties such as st-connectivity and
bipartiteness.
Other quantum algorithms provide larger subroutines for algorithms targeting specific application areas, such as
machine learning, optimization, and quantum simulation. For machine learning and optimization, examples include
variants of the quantum linear systems algorithm (QLSA) [44], quantum singular value decomposition (QSVD) [63],
quantum backtracking [64], and a quantum dynamic programming method augmented by the amplitude amplification
and estimation framework [65], among many others. Especially relevant to quantum simulation is phase estima-
tion [66], which provides a key subroutine for estimating the energy of molecular ground states (among other quanti-
ties of quantum mechanical systems) via the estimation of the eigenvalues of Hamiltonian operators. For both quantum
primitives and subroutines, the Quantum Algorithm Zoo provides an extensive catalog of quantum algorithms, their
bounds, and their formal applications [67].
Recently, device constraints have inspired the development of a wide variety of heuristic approaches for opti-
mization, machine learning, and quantum simulation with the requisite flexibility to accommodate NISQ hardware.
Among these heuristic methods targeting optimization problems are the quantum approximate optimization algorithm
(QAOA) [19] (and its generalization to the quantum alternating operator ansatz [68]) and quantum unconstrained
binary optimization (QUBO) [69], the latter of which is principally deployed by quantum annealers to target NP-
hard combinatorial optimization problems. Another class of methods, known as variational quantum algorithms
(VQAs) [70] for their use of the variational principle [71], are commonly applied for machine learning and simu-
lation tasks. Within quantum simulation, the variational quantum eigensolver (VQE) [72] has been used to target a
wide variety of problems, including estimations of electronic ground state [73], molecular binding [74] and dissocia-
tion [75], and chemical reactions [76]. More generally, variational quantum machine learning algorithms, including a
wide variety of quantum neural networks (QNNs) [77, 78, 79, 80, 81] and quantum kernel methods [82, 23], have been
applied to a wide variety of supervised, unsupervised, generative, and reinforcement learning tasks. While substantial
focus has been placed on quantum methods for supervised learning in the context of medical diagnostics, there is
growing interest in the use of generative methods to explore the chemical space of small molecules [11], which may
have relevance to drug discovery.
3.3 What are the applications that are amenable for the shift from Classical to
Quantum?
Within the scope of biomedical research, practical quantum advantages represent the key motivation for shifting from
classical to quantum methods. Past work has defined a quantum advantage as any quantum computation that improves
upon the state-of-the-art classical computation given a specific evidence context [2]. Crucially, this definition requires
the identification of i) a specific problem (or problem instance), ii) a quantum and classical algorithm pair, iii) a com-
putational resource (e.g. time, space, information), iv) bounds on that resource, and v) the specific evidence context
(i.e. theoretical, experimental, operational). By encouraging the identification of these constituent components, this
framework facilitates reasoning around how quantum advantages may be applied to real-world use cases in biomedical
research. For the biomedical research community, empirical (i.e. experimental and operational) advantages in time
and information resources are of principal interest.
Broad application areas where empirical advantages may be realized over the near- to medium-term include quan-
tum simulation and machine learning. For quantum simulation, specific advantages are expected to be experimental
and relative to time resources (up to a superpolynomial advantage) and reductions in approximation error for physical
and chemical quantities (e.g. molecular ground state energy, binding constants). Use cases where these advantages
have the potential to translate into realized operational advantages include drug discovery (e.g. compound screen-

10
ing and lead optimization) [83, 84, 85] and precision medicine. For quantum machine learning, specific advantages
are expected to be relative information resources (i.e. lower model generalization error and faster learning by fewer
examples, which is relevant to small data contexts [2]) and may readily translate from experimental to operational ad-
vantages for certain use cases. While advantages in quantum simulation are supported by robust theoretical evidence
the set of use cases is relatively narrow. In contrast, theoretical evidence supporting quantum machine learning advan-
tages remains relatively limited, especially for classical data. Nonetheless, the methodological flexibility of quantum
machine learning methods may offer a substantially broader potential scope of application.
Expanding the scope of quantum computing in biomedical research is expected to be contingent on interdisci-
plinary efforts to identify precise use cases (using the above framework) that are ripe for the development of experi-
mental and operational advantages. While quantum simulation and quantum machine learning methods are expected
to be productive areas of exploration, applications using quantum optimization methods merit exploration. Finally,
given the early stage of quantum hardware development, it should be expected that identified use cases will continue
to be highly sensitive to hardware capabilities for the foreseeable future. New devices may quickly realize new ca-
pabilities and, from a process perspective, this dynamism strongly implies that continuous evaluation identified use
cases and the feasibility of hardware-specific proof of principles may foster the growth of interdisciplinary efforts.
3.4 Existing Collaborations and Projects in the Field
Various institutions and companies have been exploring the potential of quantum algorithms in analyzing biomed-
ical data. These efforts generally aim to leverage the computational capabilities of quantum computers to address
biomedical challenges, from drug discovery to genomics.
For example, some pharmaceutical companies, in collaboration with quantum computing companies, have begun
investigating quantum algorithms for molecular modeling and drug discovery. This is due to the exponential complex-
ity of simulating molecular interactions on classical computers. More specifically, AstraZeneca has reported exploring
quantum computing to speed up structural chemistry to find "molecules that matter", given that a key element of drug
discovery and development relies on establishing the chemical 3D-structure of potential new medicines [86]. More
recently, the Alphabet spinoff SandboxAQ has signed up AstraZeneca as users of its drug discovery and development
tools [87].
As another concrete example, IBM and the Cleveland Clinic have established a landmark partnership to advance
healthcare research using AI, quantum computing, and cloud computing technology. As part of the 10-year partner-
ship, IBM installed its first private sector, on-premises IBM Quantum System One in the United States, in Cleveland.
This quantum system was designed to leverage quantum computing’s potential in aiding with calculations related to
drug discovery and other complex matters that classical computers struggle with [88]. The partnership also involves
the establishment of the Discovery Accelerator, a joint Cleveland Clinic and IBM center designed to advance research
into critical areas such as genomics, single-cell transcriptomics, population health, clinical applications, and chemical
and drug discovery [89].
Not surprisingly, the few examples of existing projects and collaborations such as the ones described above are
taking advantage of the most successful and feasible quantum approaches already being used in other scientific areas,
such as quantum simulations and hybrid quantum-classical algorithms. Consequently, there is a great opportunity
to leverage the combined expertise and research coming from both DOE and NIH to explore other ways to tackle
biomedical data (classical and quantum input) problems.

11
4. Biomedical Use-Cases for Short-Term Quantum Applicability and
Impact

4.1 Medical Imaging and Diagnostics

The field of medical imaging could greatly benefit from quantum algorithms as traditional methods for image analysis
(e.g., segmentation), reconstruction (e.g., iterative methods), and registration (e.g, feature/intensity/distance- based
methods) are computationally demanding, and the need for improved diagnostic accuracy is paramount. Quantum
algorithms, including quantum Fourier transform [3], quantum phase estimation, quantum optimization methods [4],
quantum machine learning (QML) [5], and quantum image processing techniques [6], can be adapted to enhance
medical imaging applications. The demand for quicker, more accurate, and efficient diagnostics is high, and specific
applications, such as image denoising and filtering applied to smaller scale classical and quantum input, may be
feasible with near-term quantum hardware due to their modest qubit requirements, especially the initial study and
development of proof-of-principle demonstrations that can also reduce noise to manageable levels.
Based on the discussions of the roundtable, we discuss the two main promising areas where the development
of quantum algorithms may play an important role to overcome the high computational demand of some specific
processing steps.
4.1.1 Basic Image Analysis Operations
Quantum Fourier Transform (QFT). QFT [60, 3] is a linear transformation on quantum bits and it is an essential
component of many quantum algorithms. It is a quantum version of the classical discrete Fourier transform (DFT),
which is used widely in image processing, especially for tasks like filtering, compression, and noise reduction. QFT
could potentially offer exponential speedup in such operations [90, 91]. QFT and its variations [92] and also simular
transforms such as the quantum wavelet transform [93, 94] could potentially help in the context of medical imaging
and diagnostics in terms of:

• Speed: These transformations can theoretically be performed much faster on a quantum computer than their
classical counterparts on a classical computer. This speedup could lead to quicker analysis and diagnostics,
especially in scenarios where real-time or near-real-time image analysis is crucial, such as during surgery.

• Efficiency: The effective implementation of these transforms for image analysis could potentially enable the
processing of large and complex images more efficiently. This can be particularly useful in application using
high-resolution 3D images, like those produced by an MRI or CT scan.

• Detail: By processing images in the frequency domain, it may be possible to isolate and analyze fine details in
images that would be challenging to detect in the spatial domain. This could lead to more accurate diagnoses.

Quantum Machine Learning (QML). QML algorithms and quantum classifiers can potentially be used to classify
and recognize features in medical images. For example, just like their classical counterparts, QCNNs [24] are a type
of quantum machine learning model that can be used for image analysis tasks, like feature extraction and pattern
recognition, potentially providing quantum speedups. This could significantly improve the diagnostic process and
potentially lead to earlier detection of diseases, for example. QSVMs, which are based on the optimized binary
classifier support vector machine (SVM), can be highly efficient when implemented on a quantum computer [22, 95],
enabling the possibility of their application to biomedical data classification, with great potential to help with tasks
such as the classification and recognition of features in medical images.
Quantum Image Processing (QImP). QImP techniques [96, 97] encode image information into quantum states,
allowing operations such as convolution and edge detection to be performed with quantum speedups. For instance,
tasks such as image denoising [98], edge detection [99], image segmentation [100] could potentially benefit from
these techniques. Moreover, QImP techniques can not only help to improve image quality and extract features, but
can also facilitate efficient image storage and retrieval [101]. There is also an interesting interplay between QML
and QImP techniques. In the context of biomedical research, QML could be used to analyze and interpret the output

12
of a QImP algorithm. For example, a QML algorithm could be used to classify or detect features in images that
have been processed using QImP techniques. Similarly, QML could be used to optimize the parameters of a QImP
algorithm. Conversely, QImP could provide pre-processed, high-quality image that could enhance the performance of
QML algorithms.
4.1.2 Optimization
Image Registration. Image registration is a process that aligns different sets of data into one coordinate system. In
the biomedical field, it is a crucial step in many image analysis tasks, such as comparing patient scans over time or
combining data from different imaging modalities. With quantum computing, the image registration process could
potentially be accelerated, owing to the inherent parallelism and computational capabilities of quantum systems.
This acceleration could be particularly impactful when working with large, high-resolution 3D images, which are
commonplace in the biomedical domain.
Quantum optimization algorithms that could potentially be used for image registration include the Quantum Ap-
proximate Optimization Algorithm (QAOA) [19], Variational Quantum Eigensolver (VQE) [72], and Quantum Am-
plitude Estimation (QAE), or more specifically quantum amplitude amplification and estimation [62, 65]. The QAOA
can be used for solving combinatorial optimization problems. In image registration, it could be used to find the trans-
formation that best aligns two images, an optimization problem. However, this approach would require formulating
the image registration problem in a way that is compatible with QAOA. VQE is a hybrid quantum-classical algorithm
used to find the ground state of a quantum mechanical system. It uses a parameterized quantum circuit, the parameters
of which are optimized using a classical optimizer to find the lowest eigenvalue (the ground state energy) of a given
Hamiltonian. This process of variational optimization could be adapted to tackle image registration problems. QAE is
used to estimate the probability of an outcome in a quantum mechanical system, which can be related to the objective
function in an optimization problem. While not an optimization algorithm itself, it could form part of a quantum-based
image registration process, estimating the probability of a correct alignment.
Other quantum algorithms could potentially be adapted of developed for image registration including QFT, quan-
tum phase estimation (QPE) [66], and QML (using QSVM and quantum principal component analysis (QPCA) [102]
to extract features and make the registration process more efficient and robust to variations in the images).
Image Reconstruction. Quantum computers hold potential for advancing image reconstruction methods, which are
crucial when dealing with different medical imaging modalities such as CT and MRI. The primary challenge in image
reconstruction is to recover an original image from degraded or incomplete data. Quantum algorithms could offer
speed-ups and enhanced capabilities for such tasks. Quantum algorithms mentioned before, QAOA and VQE, could
also be used in this context. In image reconstruction, where the task can be cast as the optimization problem of finding
the most likely image given the observed data, QAOA might be applied. VQE, although used for finding the ground
state of a quantum system, can also be adapted for image reconstruction tasks, where the goal would be to minimize
the difference between the reconstructed image and the observed data.

It is important to note that this use-case is rapidly evolving with the advance of quantum sensors, as described in [103],
given that better reconstruction is expected because of the higher sensitivity compared to classical sensors. Moreover,
in this case, quantum algorithms would be applied directly to the quantum data coming from these sensors, which
would avoid the data encoding step (see Section 5.1).
4.2 Biological Sequence Data
Biological sequence analysis involves the characterization and interpretation of sequences of biological molecules,
particularly DNA, RNA, and protein sequences to understand their structure, function, and evolution. This encom-
passes sequence alignment, variant/mutation detection, sequence queries, phylogenetic analysis, functional inference
and annotation, structural prediction, motif finding, comparative genomics and metagnomics among others. Given the
wide scope of potential applications [104, 2, 105], biological sequencing data offers a large sandbox for the devel-
opment of novel quantum algorithms and quantum computing methods over the near term that may provide greater
computational efficiency.
Multiple examples of novel quantum algorithms inspired by biological sequencing data already exist. For exam-
ple, with regard to biological sequence alignment and pattern matching, modifications of Grover’s algorithm have

13
been able to address the problem of repeated sequences [106]. Further extensions of Grover’s search algorithm al-
lowed approximate matches needed for read errors in genomics, and a distributed search for multiple solutions over
the quantum encoding of DNA sequences gave a quadratic speedup over the classical algorithm [107]. With respect
to phylogenetics, quantum simulation of phylogenetic inference and evolution [108] has been developed leveraging
quantum walks [109]. In metagenomics, de novo genome assembly is complicated by the variable abundance of
species within the sample and modularity optimization in community detection is an NP-hard problem. The applica-
tion of quantum annealing to de novo genome assembly [110] as well as assessment of empirical run-time of quantum
algorithms for community detection [111] provide potential avenues to address these challenges in metagenomics.
4.3 Drug Discovery and Design
Drug discovery and design was seen as a compelling use case for quantum computing due to its potential to signifi-
cantly accelerate the drug discovery process and address current complex computational challenges. Drug discovery
involves modeling and simulating complex molecular interactions between potential drug candidates and target pro-
teins. These simulations require extensive computational resources, making them an ideal candidate for quantum
acceleration. Quantum computers could also provide exponential speedup for key drug discovery tasks, such as
molecular dynamics simulations, conformational searching, and quantum chemistry calculations, leading to faster
identification of drug candidates
Hybrid quantum-classical workflows hold tremendous near-term promise. One example discussed paired classical
docking and molecular dynamics with quantum machine learning for finding ligands binding to proteins (accounting
for the presence of mutations [13]. A critical step was the identification of the components of the workflow amenable
for quantum approaches (in this example, the mutation-impact predictor). The classical machine-learning was mapped
to quantum using a neural network constructed from qubit-rotation gates and in both simulation and physical applica-
tion showed promise compared with classical approaches.
A key challenge in the drug discovery process is that only a small fraction of drug-like molecules are therapeu-
tically relevant [112, 113]. A critical step is hit identification, which involves the identification of small molecules
with adequate activity for a specific target that could then be used as a starting point for optimization. Identifying
new molecules with a specific desired biological activity, while already challenging is even more difficult when no
information about the target is known (e.g, novel target families, orphan targets etc). Molecular de novo design and
compound optimization have been aided by artificial intelligence and generative models. Quantum GANs (QuGANs)
can improve performance compared with classical GANs through more efficient search in exponentially large chem-
ical space. Scalable quantum generative autoencoder (SQ-VAE) can support simultaneously reconstructing and sam-
pling drug molecules. Comparison with classical counterparts have shown high-dimension molecules generated from
quantum generative autoencoders have better drug properties within the same learning period [11].

14
5. Discussed Cross-Cutting Supporting Research Topics
5.1 Data Preprocessing, Encoding/Decoding, and Input/Output Data
Using quantum computers to process and learn from classical data naturally requires as a first step mapping the
classical data (e.g., a medical image) to a quantum state. This encoding step is crucial to guaranteeing the success of
the quantum algorithm aimed at processing the classical information. Several methods exist to embed classical data
into quantum states such as basis encoding, angle encoding, amplitude encoding, block encoding, and IQP encoding
among others [23, 114]. Recently, an efficient method (QPIXL) for encoding classical data has been developed
tailored to NISQ devices and capable of encoding different types of data [18]. More interestingly, extensions of
QPIXL were developed later and used to develop proof-of-concept analysis algorithms to process images, time-series,
and DNA sequences on real NISQ hardware [8]. Due to the limitation of NISQ quantum computers there is usually
an intrinsic mismatch between the size of the classical data (i.e., the number of pixels in a image) and the available
quantum resources (i.e., encode every pixel value in qubit-qubit interactions). This leads to the requirement of data
downsampling methods, where the dimensionality of the classical data is reduced via classical post-processing and
only later embedded into a quantum state. Data reduction methods include principal component analysis (PCA),
autoencoding, feature learning or feature extractions. While using these techniques can enable quantum computers to
process large classical datum, there are several limitations that must be accounted for. On the one side there are data
compression limitations one must consider. Similarly, the efficiency of the data reduction scheme must be accounted
for in the computational paradigm. More importantly, if one is to understand what are the capabilities and limitations
of using quantum computers to process classical data, the use of data reduction protocols raises an important question:
What is the true role of the quantum device given that a classical computer has already done a significant and crucial
portion of the analysis? From a theoretical perspective, it will be crucial to understand and answer these sorts of
questions.
A different approach to using near-term quantum computers is to analyze the so-called quantum data. Quantum
data refers to quantum states encoding some valuable information which are obtained by some quantum mechanical
process. In the context of biomedical research one can think about using quantum mechanical systems, allowing them
to interact with their environment, and learn something from the environment through this interaction. This sort of
approach has been recently identified as one of the potential use-cases of near-term quantum technologies [115].
5.2 Quantum Optimization
Quantum optimization refers to the use of quantum algorithms and quantum computers to solve optimization problems.
Optimization problems are pervasive in many fields and typically involve finding the best solution (maximum or
minimum) from a set of feasible solutions based on a certain criteria. As mentioned in other sections of this report,
quantum optimization methods can have an important impact when developing quantum algorithms for biomedical
research. In this section, we briefly summarize some of the existing quantum optimization algorithms that can be
used/adapted for the purposes of quantum biomedical applications.
Quantum Annealing (QA): QA is a quantum optimization technique inspired by the classical optimization method
called simulated annealing. It is primarily used for solving complex combinatorial optimization problems. The clas-
sical simulated annealing is inspired by the annealing process in metallurgy, which involves heating a material and
then allowing it to cool slowly to remove defects and finds a low-energy state. The algorithm similarly tries to find
a minimum (or maximum) solution to a problem by exploring the solution space, starting with a high “temperature”
(meaning solutions can widely vary) and gradually “cooling” to converge on an optimal or near-optimal solution.
Instead of temperature, QA uses quantum effects, such as tunneling for example, to explore the solution space. Quan-
tum tunneling allows the system to jump directly through energy barriers rather than climbing over them, potentially
speeding up the search for an optimal solution. The goal of QA is to find the ground state (the lowest energy state) of
a quantum system, which corresponds to the optimal solution of the optimization problem. QA requires specialized
hardware, a well-defined problem (for efficient convergence), and is not typically executed on gate-based quantum
computers. There is ongoing debate and research about how much speedup D-Wave’s annealers provide over classical
algorithms for various optimization problems. Nevertheless, QA has the potential to significantly impact biomedical
research by addressing some well-known optimization problems including protein folding, drug discovery (identify
drug candidates), genomic data analysis (feature selection, clustering, etc), medical imaging (image reconstruction,

15
segmentation, and registration), optimizing treatment plans, and epidemiological models.
Quantum Approximate Optimization Algorithm (QAOA): QAOA is a heuristic quantum algorithm developed for
tackling combinatorial optimization problems, and it is one of the promising techniques in the noisy intermediate-
scale quantum (NISQ) era. This hybrid quantum-classical algorithm is a type of variational quantum algorithm, i.e., it
iteratively refines a quantum state by adjusting a set of parameters to minimize the expectation value of a Hamiltonian
(representation of the problem in question). Compared with QA, both QAOA and QA aim to find low-energy states of
problem Hamiltonians, but they approach this task differently. QA relies on the physical process of annealing while
QAOA uses a more algorithmic, gate-based approach. Relevant to the NISQ era, QAOA that work with shorter, shal-
lower circuits are of particular interest. QAOA can be applied to the same optimization problems mentioned before
for QA, however the way they would be implemented and their suitability for specific problems within the domains
mentioned might differ. QAOA can be specifically interesting for problems including combinatorial design of thera-
pies, parameter tuning in bioinformatics tools, dynamic systems optimization, optimal experimental design, network
analysis, and pathway analysis in genomics. It is worth mentioning also the Quantum Alternating Operator Ansatz,
which is an extension of QAOA that generalizes the QAOA algorithms to address broader optimization problems.
Grover’s Algorithm: While not strictly an optimization algorithm, Grover’s provides a quadratic speedup for un-
structured search problems and it can be adapted for certain optimization tasks. Specifically for biomedical research,
Grover’s can be relevant for searching problems in a binary solution space, cost function evaluation, amplifying
promising solutions, combinatorial optimization, quantum-enhanced metaheuristics, and hybrid approaches.
Variational Quantum Eigensolver (VQE): VQE is an heuristic quantum-classical algorithm that was primarily
designed to determine the ground state energy of a given molecular Hamiltonian. It aims to approximate the ground
state energy of a quantum system, particularly molecules in the realm of quantum chemistry. VQE can certainly be
adapted and applied to various problems in biomedical data analysis including drug discovery and design, molecular
dynamics simulations, optimization problems in genomics, structural biology, and pattern recognition in biomedical
imaging.
Quantum Random Walks (QRW): QRWs are the quantum analog of classical quantum random walks. Instead
of a definite path determined by random events (like coin tosses in classical random walks), QRWs use quantum
superposition and interference, allowing for simultaneous exploration of multiple paths. This leads to distinctively
different behaviors compared to classical walks, with potential applications in quantum algorithms and information
processing. In biomedical data analysis, QRWs have the potential to be applied to problems such as drug discovery,
genomic data analysis, protein folding, medical imaging, and epidemiological models.
5.3 Machine Learning and Neural Networks in Biomedical Research
In the near-term, one of the most promising methods for analyzing data embedded in quantum states is through Quan-
tum Machine Learning (QML). QML attempts to leverage quantum computers by adapting a hybrid computational
approach where quantum computers are used to estimate (classically hard) quantities, while the power of classical
optimizers is leveraged to train the QML model parameters. By pushing some of the computational complexity onto
classical devices, QML aims to reduce the requirements on the quantum hardware, making it more amenable for
present-day technologies.
By utilizing the Hilbert space of quantum states as a quantum-enhanced feature space [23, 82] QML is able to
generalize many classical machine learning methods to the quantum realm. For instance, quantum computers can be
used to estimate similarity measures between quantum states, and thus construct a kernel which can be used in kernel-
based techniques such as support vector machines (SVM) [23]. Similarly, quantum versions of several quantum neural
networks (QNNS) have been proposed and developed [77, 78, 79, 80, 81]. At their core, a quantum and a classical
neural network perform a similar role. They take input data, which is processed through a series of layers, and finally
mapped to some output space. Despite this similarity, their modus operandi are drastically different. Classical neural
networks rely greatly on non-linearities and the ability to copy classical data. On the other hand, quantum learning
models are constrained by the no-cloning theorem [116] and quantum neural networks are usually instantiated via
parametrized quantum circuits meaning that they can only act linearly, and unitarily, on the input data. As such, non-
linearities in QML are usually implemented at the data-embedding level, through data re-uploading schemes [114], or
by accessing multiple copies of the input data states [117, 118].

16
 A crucial aspect to note is that while deep learning has gained significant attention in classical machine learning, its
quantum counterpart has several limitations that must be accounted for. For instance, the absence of backpropagation
methods for QNNs makes gradient-based training of quantum models computationally expensive, as every single
gradient entry must be individually computed. Moreover, deep quantum circuits are prone to exhibit barren plateaus
(i.e., vanishing gradients) [119, 120] or large amounts of local minima in their parameter training landscape [121].
Despite such limitations QNNs have shown promise to outperform their classical counterparts [79], and more work is
needed to understand their capabilities in domain specific applications such as biomedical research. Specifically, it has
been recently shown that by creating QNNs with sharp inductive biases one can mitigate issues and lead to efficiently
trainable models [80].
5.4 Foundational Mathematical Concepts
Several foundational mathematical concepts underpin the development of new quantum algorithms for biomedical
data analysis. Here are some of them:

• Linear Algebra: Quantum states are described using vectors, and quantum operations are represented by matri-
ces. So understanding vector spaces, matrix operations, eigenvalues, and eigenvectors is crucial. Linear algebra
is the mathematical language of quantum mechanics and quantum computing.

• Probability Theory: Quantum mechanics is inherently probabilistic. When we measure a quantum state, we get
one of several possible outcomes, each with a different probability. Understanding the theory of probability is
necessary to make sense of this.

• Fourier Analysis: Quantum computing uses Fourier analysis in many algorithms, like Shor’s algorithm or the
Quantum Fourier Transform. Understanding the basics of Fourier series and the Fourier transform is critical.

• Differential Equations: Quantum systems evolve according to the Schrödinger equation, which is a differential
equation. Having a solid grasp of how to solve and analyze such equations is beneficial.

• Graph Theory: Some quantum algorithms, like the quantum walk, are based on graph theory. Moreover, many
problems in biomedical data analysis, such as understanding genetic networks or the structure of chemical
compounds, can be framed as problems in graph theory.

• Group Theory: Quantum computing uses group theory, especially in algorithms for solving the hidden subgroup
problem. This is an important mathematical concept in the field.

• Sampling: Sampling, the process of extracting classical information from quantum states, is crucial in quantum
computing. Used in algorithms like Shor’s and QFT, it leverages quantum superposition to generate probabilistic
results. In biomedical applications, it can aid in generating complex models, simulating random processes, and
solving optimization problems.

These mathematical tools, among others, form the building blocks for developing and understanding quantum
algorithms for biomedical data analysis. They provide the language and framework needed to describe and manipulate
quantum states, design quantum circuits, and analyze their behavior. Table 5.1 shows the intersection between these
foundational mathematical concepts and the quantum algorithms and applications discussed in this report.
5.5 Addressing Noise
When it comes to the development of machine learning algorithms, considering noise associated with the data is an
important part of the process in a number of ways. Sample noise can help to improve the robustness of models by
making them less sensitive to small changes in the data. Intentionally added noise can be used to regularize models and
prevent overfitting. Unwanted noise can be mitigated through techniques such as data cleaning and feature selection.
Sample noise refers to random or inherent variability in the data, which can significantly impact the performance
of algorithms. It often leads to reduced model accuracy due to overfitting, where the model learns noise as if it were a
significant part of the data, resulting in poor generalization to new data. Sample noise, although often challenging, can
be beneficial by promoting model robustness and preventing overfitting. It encourages models to focus on significant

17
Mathematical Concept Quantum Algorithms Potential Applications
Linear algebra Quantum Phase Estimation (QPE), Quan- Medical Imaging, Molecular Dynamics, Pat-
tum Fourier Transform (QFT), HHL Algo- tern Recognition, Protein Folding, Drug De-
rithm, Quantum Singular Value Decomposition sign, Phylogenetic Trees
(QSVD), Shor’s Algorithm, Variational Quan-
tum Eigensolver (QVE), Swap Test, Amplitude
Amplification
Probability Theory Grover’s Algorithm, Quantum Phase Estima- Drug Interactions, Drug Discovery, Molecular
tion (QPE), Quantum Counting, Amplitude Es- Dynamics, Protein Folding, Pattern Recogni-
timation, HHL Algorithm, Quantum Random tion, Disease Classification, Drug Target Iden-
Walks, Quantum Machine Learning, Swap Test tification
Fourier Analysis Quantum Fourier Transform (QFT), Shor’s Al- Medical Imaging, Genomic Data Analysis,
gorithm, Quantum Phase Estimation, HHL Al- Protein Structure Estimation, Drug Interac-
gorithm, Simon’s Algorithm, Quantum Walks tions, Linear Systems in Biomedical Model-
in the Fourier Basis ing, Biomarker Discovery, Biological Systems
Modeling
Differential Equations Quantum Simulation Algorithms, HHL Algo- Epidemiological/Neurological/Cardiovascular
rithm, Quantum Machine Learning, Adiabatic Modeling, Protein Folding, Genomic Data
Quantum Computing and Quantum Annealing Analysis, Image Analysis, Predictive Model-
ing, Systems Biology
Graph Theory Quantum Walks on Graphs, Quantum Search Genetic Linkage Maps, Protein-Protein Interac-
on Structured Data, Quantum Algorithms tion Networks, Disease Modeling, Brain Con-
for Graph Problems, Quantum Approxima- nectivity Networks, Medical Imaging, Metage-
tion Algorithms, Adiabatic Quantum Comput- nomic Data Analysis, Protein Structure Predic-
ing and Quantum Annealing, Quantum Error- tion
Correction, Graph States, Quantum Algorithms
for Shortest Path and Related Problems
Group Theory Shor’s Algorithms, Simon’s Algorithm, Quan- Protein Folding, Molecular Symmetry Analy-
tum Phase Estimation (QPE), Hidden Subgroup sis, Genomic Data Analysis, Medical Imag-
Problem (HSP), Quantum Algorithms for Non- ing, Drug Interaction and Design, Phylogenetic
Abelian Hidden Subgroup Problems, Jones Trees
Polynomial and Topological Quantum Com-
putation, Quantum Walks on Cayley Graphs,
Solovay-Kitaev Algorithm
Sampling Quantum Fourier Transform (QFT), Quantum Drug Discovery, Genomic Data Analysis,
Amplitude Estimation (QAE), Quantum Ap- Adaptive Trial Design, Disease Spread Sam-
proximate Counting, Boson Sampling, Random pling, Proteome Sampling, Statistical Model
Circuit Sampling, HHL Algorithm, Variational Sampling, Evolutionary Pathway Sampling,
Quantum Eigensolver (VQE), Quantum Ran- Behavioral Pattern Sampling, Medical Imaging
dom Walks

Table 5.1

18
patterns rather than minor fluctuations, enhancing their generalization to real-world data. Training on noisy data
also helps in simulating practical scenarios and can lead to more sophisticated feature selection and realistic model
validation.
To prevent overfitting, which occurs when a model learns the training data too well and does not generalize well
to new data, intentionally added noise forces the model to learn the underlying signal instead of the noise. This can be
done by adding Gaussian noise, salt and pepper noise, or other types of noise to the data.
Unwanted noise refers to any random, irrelevant, or extraneous information or disturbances that corrupt the signal
or data. It is something that is not part of the intended signal or data and is usually considered detrimental. This type of
noise can be caused by a number of things, such as sensor noise, measurement error, and random fluctuations. There
are a number of techniques that can be used to mitigate unwanted noise, such as data cleaning and feature selection.
Data cleaning is the process of removing or correcting errors in the data. Feature selection is the process of selecting
the most relevant features for the model.
In summary, noise can be both a positive and negative factor in machine learning. It is important to understand the
different types of noise and how they can affect machine learning models, which can highly impact the development
of new quantum machine learning algorithms for biological problems. Specifically, considering these types of noises
when encoding classical data becomes crucial.
In addition to noise associated to the data, the noise found in quantum systems is equally important. Fully un-
derstanding realistic noise in quantum computers can help with tailoring specific problems in machine learning for
biological problems to the underlying noise inherent to quantum systems. More specifically, it would enable the devel-
opment of more accurate and noise-resistant quantum algorithms, significantly enhancing the efficiency and reliability
of machine learning models for complex biological data analysis. Such advancements could help advancing fields
like genomics and personalized medicine by enabling more sophisticated data processing and handling of larger, more
complex datasets with greater fidelity.
It is crucial to remember that realistic noise limits the simulable regime of quantum algorithms by reducing the
coherence time and introducing errors, which constrain the duration, complexity, and accuracy of simulations. This
limitation restricts the scalability of quantum algorithms, defining the current boundaries of feasible quantum compu-
tations (under 50 qubits) and the range of problems that can be effectively addressed. To support the development of
quantum algorithms taking into account noise, several noise models can be simulated such as, for example, decoher-
ence, dephasing, crosstalk, low frequency, among others.
Some quantum algorithms explicitly take advantage of the noisy nature of the underlying quantum hardware for
the design of noise-robust algorithms or error mitigation methods [122]. But a similar attempt has yet to be made in
the domain of machine learning and biological applications.
5.6 Privacy Preserving
Current challenges in biomedical data privacy. Modern biomedical databases are mostly massive and include
complex data from multiple sources incorporating both phenotype and genotype (e.g., biobank data, EMRs, and
general behavior data). These databases are continuously updated, may be retained for indefinite periods of time, and
can serve in unlimited research endeavors. This modern context presents multiple challenges to the classical notions
of “consent or de-identify”: (i) Scholars have cast doubts on the ability of participants to exert control in complex big
data scenarios (leading to a rise in non-informed forms of consent such as broad and opt out consents), additionally,
(ii) the detailed participants’ data available in these modern databases compounded by the rise in publicly available
information (online) made it possible to re-identify what was thought to be properly de-identified data. As a result
data curators are moving away from data sharing in favor of more restrictive and virtual data access. Current research
directions in this area include:

1. An increasingly popular way to overcome the issues of data privacy and availability is to use fully synthetic data.
The application of synthetic data to the biomedical domain is recent yet increasing in popularity. Synthetic data
is still in the experimental stage and is often used to carry out preliminary research such as exploratory analyses
as their utility is not well established. Current research is exploring new synthetic data generation mechanisms
based in machine learning as well as quantifying the utility and privacy of such generators in real-life scenarios.

2. Another research direction is the use of federated learning, which allows multiple data curators to collectively

19
 build high-quality machine learning models on their distributed data while keeping their raw data private. FL
suffers from multiple shortcomings, they require intense communication between the data holders to achieve the
final model, they rely on heavy cryptographic protocols (secure communication) to achieve results with good
accuracy while maintaining privacy, and the accuracy of the resulting models is dependent on the original data
distributions.

Quantum research directions. Privacy research can greatly benefit from the promise of quantum computing. Ex-
amples include: the exploration of more efficient secure communications, designing efficient encryptions that enable
analysis on encrypted data and designing superior synthetic data generators:

1. Encryption is used to secure communications in federated learning models. It is preferred over other privacy
preserving mechanisms as it does not distort intermediate parameter calculations. The impact of quantum
machines on the efficiency of secure communications is something to anticipate, provided a quantum secure
encryption is used.

2. Fully homomorphic encryption (FHE) is an ideal tool for the private analysis of sensitive data. It allows any
analysis (any function) to be applied on encrypted data while providing the same encrypted results if the analysis
were to be applied on plain data (imagine having experts analyze your genomic data without obtaining any
information about the data or the results). Fully homomorphic encryption is not in use due to its high inefficiency
in real life scenarios. Quantum computing could be explored to:

(a) Study the impact of quantum machines on the efficiency and security of existing FHE’s
(b) Generate novel efficient FHEs
(c) On the Synthetic data generation front, an interesting direction is to explore the use of quantum machine
learning to design SDGs of higher efficiency and utility

5.7 Quantum Simulation and Real Quantum Hardware

Understanding where, and how, quantum technologies can be used to enhance biomedical computations will likely
require large amounts of benchmarking and model testing. At a first stage, one can simulate the quantum algorithm
via classical methods to understand what are its capabilities in an ideal setting. Techniques such as tensor networks,
or sparse matrix calculations allow for the simulation of generic quantum computation for systems of up to 20 or 30
qubits in standard-sized machines (larger simulations are possible under certain assumptions). These simulations can
be used to benchmark the algorithm’s performance in a noiseless setting (i.e., no hardware noise, and no sampling
noise), and thus study its performance without other sources of uncertainty. Here, it will be crucial to leverage these
early simulations to study the scalability of the quantum algorithm. That is, in order to guarantee that the methods
will perform well in realistic problems, one should perform analysis of the number of resources (e.g., number of
training iterations) needed to achieve a given solution accuracy as a function of the problem size. Such study will
guarantee that the quantum algorithm complexity will not scale prohibitively for large scale problems. At a second
stage, noise should be included in these classical simulations. Since near-term quantum hardware is intrinsically noisy,
algorithms aimed at NISQ computers should be studied with sampling and hardware noise. Including sampling noise
in a simulation allows one to study the number of calls to a quantum computer needed to solve a given task. Crucially,
by adding hardware noise one can understand the robustness of the algorithm, as well as the need to integrate error
mitigation techniques along with the algorithm itself.
Once the aforementioned classical simulations have been performed, one can use quantum hardware to test in a
real device the performance of the quantum algorithm. Given that different platforms exist –superconducting qubits,
trapped ions, neutral atoms, photonics-based– the practitioner should identify what is the device that best suits its
algorithm needs. Here there are several factors that come into play and that must be accounted for, such as the number
of qubits, the gate error rates, and the qubit topology and connectivity. In all cases, the classical benchmarking
of simulations will be crucial in estimating the performance of the algorithms, as well as its run cost (i.e., how many
shots are needed, or how much quantum computer run-time is needed). While real quantum hardware implementations
are the ultimate goal, current devices are still in development stages. Even if available devices are not suited for

20
implementing a given algorithm, this does not preclude the possibility that their near-future versions will be. As such,
a useful analysis that one can perform is: how good does the hardware need to be for the algorithm to succeed? This
kind of study can be further used to co-design hardware amenable for a given task.
5.8 Workforce Development and Accessibility
The span of applications for quantum computing in biomedical data science will not be met by a single pedagogical
approach nor a single workforce persona. There is a need for a variety of expertise and education levels to create a
balanced technical workforce. In addition, the continually evolving innovations in the field pose additional challenges
for standardized curriculum, requiring a continual learning mindset for those in this field.
To date, the focus of training and workforce development has been on Quantum Information Science (QIS). In
2021, a DOE Office of Science working group developed a plan to establish the knowledge base and competencies
needed for developing a curriculum in QIS [123]. In the report, two communities were identified who should con-
tribute to curriculum development for QIS: the “demand side” (industries supporting development and manufacturing
of technologies based on QIS as well as government laboratories and universities conducting research and devel-
opment in QIS) and the “supply side.” ( educational, degree granting institutions, as well as the National Science
Foundation which directly supports educational research). While this is critical for quantum workforce development,
additional skills and competencies need to be considered for applied quantum biological data science.
Key components of such training must include interdisciplinary training (both with respect to mentoring and
pairing of trainees), hands-on training, emphasis on team science and critical thinking in order to appropriately
evaluate applications of quantum approaches in the context of biology and medicine. Accessibility in terms of QIS is
often focused on access to technology through the cloud and quantum simulators. While this is absolutely critical, in
order for the uptake of quantum computing and algorithms in biological data science to be meaningful, there must also
be a focus on cognitive accessibility. This requires to first understand the potential users of these approaches in biology
and medicine and then focus on user experience and user interface design when developing software and toolkits.
Software development is not enough - these approaches must be integrated into existing domain specific workflows
(as well as developed for those use cases). Leveraging the expertise and existing work in both NIH and DOE could
allow this to be accomplished in the near-term through joint training programs and curriculum development.
5.9 Ethics and Governance
It is clear that Quantum computing presents new ethical and governance issues that must be considered [124] . For
example, encryption and security breakdowns could lead to unprecedented security breaches and privacy violations.
Quantum sensors could enhance surveillance capabilities, raising ethical questions about privacy, and civil liberties. In
addition, access to quantum computing hardware could enable certain groups and organizations, as well as nations to
gain significant advantages in areas such as cryptography, optimization, and drug discovery leading to ethical concerns
about power concentration, inequality and access to the technology [125] .
There are also ethical and governance aspects of quantum computing that are shared with fields such as AI/ML.
These include bias and fairness, transparency and accountability, with a need for ethical governance structures to
ensure responsible development and deployment. There is also an interplay between quantum computing and AI that
must be considered. Quantum computing will enhance AI capabilities, making it essential to consider the ethics of AI
within the context of quantum computing as well.
Preparing for the ethical and governance challenges of quantum computing requires a proactive and strategic ap-
proach. We are in a critical window to develop the ethical and governance frameworks and before the full potential
of quantum is realized. While the ethical landscape is evolving as quantum technologies advance, there is much that
can be learned from the ethical issues raised by the advent of AI/ML to ensure past mistakes are not repeated. Proac-
tively development of comprehensive ethical frameworks must be undertaken to establish clear ethical principles and
guidelines. Engagement with ethicists and interdisciplinary collaboration must be encouraged to integrate ethical con-
siderations into quantum research and development processes. Robust governance structures must be established to
oversee quantum computing research and development and establish mechanisms for sharing knowledge, technology,
and insights while addressing security concerns. Transparency and inclusivity must be integral components of gov-
ernance processes, involving diverse stakeholders, including academia, industry, nonprofits and government. Finally,
there needs to be major investment in education and public awareness as well as promoting ethical literacy within or-

21
ganizations and institutions engaged in quantum research and application to encourage the responsible use of quantum
computing.
Establishing a solid ethical foundation and robust governance structures early that guides the development and
deployment of quantum computing technology is key. By proactively addressing ethical concerns, the benefits of
quantum computing can be realized while minimizing risks and ensuring that the technology serves the broader inter-
ests of society

22
6. Conclusion

In conclusion, the intersection of quantum computing, data science, and biomedical research represents an exciting
frontier with immense potential. The collaborative efforts between the Department of Energy (DOE) and the National
Institutes of Health (NIH) hold the promise of transformative breakthroughs in the near-, medium-, and long-term.
From addressing the computational challenges in medical imaging and diagnostics to advancing biological sequence
analysis and streamlining drug discovery, quantum computing holds the hope to make a significant impact. Cross-
cutting research topics such as data preprocessing, quantum algorithms, noise mitigation, privacy preservation, and
quantum hardware development further underscore the potential of this collaboration. As we embark on these research
directions, it is essential to focus on workforce development, accessibility, ethics, and governance to ensure responsible
and equitable progress.

23
7. Addendum

During the review of this report, a proof-of-principle 48 qubit fault tolerant quantum computer (FTQC) was demon-
strated with the potential to readily scale to as many as 1000 logical qubits over the near term [126]. This result gives
significant reason for optimism over the near- to medium-term and implies that the challenges due to NISQ error rates
and NISQ-only algorithmic paradigms (e.g. certain variants of QAOA) may be obviated in the coming years. For the
timing, this work may prove critical to advancing the viability of quantum computing in the biomedical data sciences
and guiding the development of near-term practical quantum advantages.

24
Bibliography

[1] Shakuntala Baichoo and Christos A. Ouzounis. Computational complexity of algorithms for sequence compar-
ison, short-read assembly and genome alignment. Biosystems, 156-157:72–85, 2017.
[2] Benjamin A. Cordier, Nicolas P. D. Sawaya, Gian Giacomo Guerreschi, and Shannon K. McWeeney. Bi-
ology and medicine in the landscape of quantum advantages. Journal of The Royal Society Interface,
19(196):20220541, 2022.
[3] Daan Camps, Roel Van Beeumen, and Chao Yang. Quantum Fourier transform revisited. Numerical Linear
Algebra with Applications, 28(1):e2331, 2021.
[4] Yangyang Li, Mengzhuo Tian, Guangyuan Liu, Cheng Peng, and Licheng Jiao. Quantum Optimization and
Quantum Learning: A Survey. IEEE Access, 8:23568–23593, 2020.
[5] Muhammad Junaid Umer and Muhammad Imran Sharif. A Comprehensive Survey on Quantum Ma-
chine Learning and Possible Applications. International Journal of E-Health and Medical Communications
(IJEHMC), 13(5):1–17, 2022.
[6] Sanjay Chakraborty, Sudhindu Bikash Mandal, and Soharab Hossain Shaikh. Quantum image processing:
challenges and future research issues. International Journal of Information Technology, 14(1):475–489, Feb
2022.
[7] Mahdi Boroujeni, Soheil Ehsani, Mohammad Ghodsi, Mohammadtaghi Hajiaghayi, and Saeed Seddighin. Ap-
proximating Edit Distance in Truly Subquadratic Time: Quantum and MapReduce. J. ACM, 68(3), may 2021.
[8] Jan Balewski, Mercy G. Amankwah, Roel Van Beeumen, E. Wes Bethel, Talita Perciano, and Daan Camps.
Quantum-parallel vectorized data encodings and computations on trapped-ions and transmons QPUs. arXiv,
2023. https://doi.org/10.48550/arXiv.2301.07841.
[9] Y. Aharonov, L. Davidovich, and N. Zagury. Quantum random walks. Phys. Rev. A, 48:1687–1690, Aug 1993.
[10] Hasan Mustafa, Sai Nandan Morapakula, Prateek Jain, and Srinjoy Ganguly. Variational Quantum Algorithms
for Chemical Simulation and Drug Discovery. 2022 International Conference on Trends in Quantum Comput-
ing and Emerging Business Technologies (TQCEBT), pages 1–8, 2022.
[11] Junde Li, Rasit O. Topaloglu, and Swaroop Ghosh. Quantum Generative Models for Small Molecule Drug
Discovery. IEEE Transactions on Quantum Engineering, 2:1–8, 2021.
[12] Kushal Batra, Kimberley M Zorn, Daniel H Foil, Eni Minerali, Victor O Gawriljuk, Thomas R Lane, and
Sean Ekins. Quantum Machine Learning Algorithms for Drug Discovery Applications. J Chem Inf Model,
61(6):2641–2647, May 2021.
[13] Bayo Lau, Prashant S Emani, Jackson Chapman, Lijing Yao, Tarsus Lam, Paul Merrill, Jonathan Warrell,
Mark B Gerstein, and Hugo Y K Lam. Insights from incorporating quantum computing into drug design
workflows. Bioinformatics, 39(1), 12 2022. btac789.
[14] Manuela Weigold, Johanna Barzen, Frank Leymann, and Marie Salm. Data Encoding Patterns for Quantum
Computing. In Proceedings of the 27th Conference on Pattern Languages of Programs, PLoP ’20, USA, 2022.
The Hillside Group.
[15] Israel F. Araujo, Daniel K. Park, Francesco Petruccione, and Adenilton J. da Silva. A divide-and-conquer
algorithm for quantum state preparation. Scientific Reports, 11(1):6329, Mar 2021.
[16] Manuela Weigold, Johanna Barzen, Frank Leymann, and Marie Salm. Expanding Data Encoding Patterns
For Quantum Algorithms. In 2021 IEEE 18th International Conference on Software Architecture Companion
(ICSA-C), pages 95–101, 2021.

25
[17] Maria Schuld, Ryan Sweke, and Johannes Jakob Meyer. Effect of data encoding on the expressive power of
variational quantum-machine-learning models. Phys. Rev. A, 103:032430, Mar 2021.

[18] Mercy G. Amankwah, Daan Camps, E. Wes Bethel, Roel Van Beeumen, and Talita Perciano. Quantum pixel
representations and compression for N-dimensional images. Scientific Reports, 12(1):7712, May 2022.

[19] Edward Farhi, Jeffrey Goldstone, and Sam Gutmann. A quantum approximate optimization algorithm. arXiv,
2014. https://doi.org/10.48550/arXiv.1411.4028.

[20] Lov K. Grover. A Fast Quantum Mechanical Algorithm for Database Search. In Proceedings of the Twenty-
Eighth Annual ACM Symposium on Theory of Computing, STOC ’96, page 212–219, New York, NY, USA,
1996. Association for Computing Machinery.

[21] Edward Farhi, Jeffrey Goldstone, Sam Gutmann, Joshua Lapan, Andrew Lundgren, and Daniel Preda. A
Quantum Adiabatic Evolution Algorithm Applied to Random Instances of an NP-Complete Problem. Science,
292(5516):472–475, 2001.

[22] Patrick Rebentrost, Masoud Mohseni, and Seth Lloyd. Quantum support vector machine for big data classifi-
cation. Phys. Rev. Lett., 113:130503, Sep 2014.

[23] Vojtěch Havlíček, Antonio D. Córcoles, Kristan Temme, Aram W. Harrow, Abhinav Kandala, Jerry M. Chow,
and Jay M. Gambetta. Supervised learning with quantum-enhanced feature spaces. Nature, 567(7747):209–
212, Mar 2019.

[24] Iris Cong, Soonwon Choi, and Mikhail D. Lukin. Quantum convolutional neural networks. Nature Physics,
15(12):1273–1278, Dec 2019.

[25] Longhan Wang, Yifan Sun, and Xiangdong Zhang. Quantum deep transfer learning. New Journal of Physics,
23(10):103010, oct 2021.

[26] Vanda Azevedo, Carla Silva, and Inês Dutra. Quantum transfer learning for breast cancer detection. Quantum
Machine Intelligence, 4(1):5, Feb 2022.

[27] Sonia Díaz-Santos and Daniel Escanez-Exposito. Classical vs. Quantum Machine Learning for Breast Cancer
Detection. In 2023 19th International Conference on the Design of Reliable Communication Networks (DRCN),
pages 1–5, 2023.

[28] Kevin Mato, Riccardo Mengoni, Daniele Ottaviani, and Gianluca Palermo. Quantum molecular unfolding.
Quantum Science and Technology, 7(3):035020, jun 2022.

[29] Jonas Landman, Natansh Mathur, Yun Yvonna Li, Martin Strahm, Skander Kazdaghli, Anupam Prakash, and
Iordanis Kerenidis. Quantum Methods for Neural Networks and Application to Medical Image Classification.
Quantum, 6:881, December 2022.

[30] Michael Ragone, Paolo Braccia, Quynh T. Nguyen, Louis Schatzki, Patrick J. Coles, Frederic Sauvage, Martin
Larocca, and M. Cerezo. Representation theory for geometric quantum machine learning, 2023. https:
//doi.org/10.48550/arXiv.2210.07980.

[31] Daan Camps and Roel Van Beeumen. Approximate quantum circuit synthesis using block encodings. Phys.
Rev. A, 102:052411, Nov 2020.

[32] Daan Camps and Roel Van Beeumen. FABLE: Fast Approximate Quantum Circuits for Block-Encodings. In
2022 IEEE International Conference on Quantum Computing and Engineering (QCE), pages 104–113, 2022.

[33] Yulong Dong, Lin Lin, and Yu Tong. Ground-state preparation and energy estimation on early fault-tolerant
quantum computers via quantum eigenvalue transformation of unitary matrices. PRX Quantum, 3:040305, Oct
2022.

26
[34] Zhiyan Ding and Lin Lin. Simultaneous estimation of multiple eigenvalues with short-depth quantum circuit
on early fault-tolerant quantum computers. arXiv, 2023. https://doi.org/10.48550/arXiv.2303.
05714.

[35] Zhiyan Ding and Lin Lin. Even shorter quantum circuit for phase estimation on early fault-tolerant quantum
computers with applications to ground-state energy estimation. PRX Quantum, 4:020331, May 2023.

[36] William M. Watkins, Samuel Yen-Chi Chen, and Shinjae Yoo. Quantum machine learning with differential
privacy. Scientific Reports, 13(1):2453, Feb 2023.

[37] Samuel Yen-Chi Chen and Shinjae Yoo. Federated Quantum Machine Learning. Entropy, 23(4), 2021.

[38] Rod Rofougaran, Shinjae Yoo, Huan-Hsin Tseng, and Samuel Yen-Chi Chen. Federated quantum machine
learning with differential privacy, 2023.

[39] Michael A. Nielsen and Isaac L. Chuang. Quantum Computation and Quantum Information: 10th Anniversary
Edition. Cambridge University Press, 2010.

[40] Leo Zhou, Sheng-Tao Wang, Soonwon Choi, Hannes Pichler, and Mikhail D. Lukin. Quantum approximate
optimization algorithm: Performance, mechanism, and implementation on near-term devices. Phys. Rev. X,
10:021067, Jun 2020.

[41] Shouvanik Chakrabarti, Andrew M. Childs, Tongyang Li, and Xiaodi Wu. Quantum algorithms and lower
bounds for convex optimization. Quantum, 4:221, January 2020.

[42] Andrew M. Childs, Jin-Peng Liu, and Aaron Ostrander. High-precision quantum algorithms for partial differ-
ential equations. Quantum, 5:574, November 2021.

[43] Dominic W. Berry, Andrew M. Childs, Aaron Ostrander, and Guoming Wang. Quantum algorithm for linear
differential equations with exponentially improved dependence on precision. Communications in Mathematical
Physics, 356(3):1057–1081, Dec 2017.

[44] Aram W. Harrow, Avinatan Hassidim, and Seth Lloyd. Quantum algorithm for linear systems of equations.
Phys. Rev. Lett., 103:150502, Oct 2009.

[45] András Gilyén, Yuan Su, Guang Hao Low, and Nathan Wiebe. Quantum singular value transformation and
beyond: Exponential improvements for quantum matrix arithmetics. In Proceedings of the 51st Annual ACM
SIGACT Symposium on Theory of Computing, STOC 2019, page 193–204, New York, NY, USA, 2019. Asso-
ciation for Computing Machinery.

[46] Adam Holmes, Sonika Johri, Gian Giacomo Guerreschi, James S Clarke, and A Y Matsuura. Impact of qubit
connectivity on quantum algorithm performance. Quantum Science and Technology, 5(2):025009, mar 2020.

[47] Kristan Temme, Sergey Bravyi, and Jay M. Gambetta. Error mitigation for short-depth quantum circuits. Phys.
Rev. Lett., 119:180509, Nov 2017.

[48] Abhinav Kandala, Kristan Temme, Antonio D. Córcoles, Antonio Mezzacapo, Jerry M. Chow, and Jay M. Gam-
betta. Error mitigation extends the computational reach of a noisy quantum processor. Nature, 567(7749):491–
495, Mar 2019.

[49] Daniel Bultrini, Max Hunter Gordon, Piotr Czarnik, Andrew Arrasmith, M. Cerezo, Patrick J. Coles, and
Lukasz Cincio. Unifying and benchmarking state-of-the-art quantum error mitigation techniques. Quantum,
7:1034, June 2023.

[50] Tian-Yu Yang, Yi-Xin Shen, Zhou-Kai Cao, and Xiang-Bin Wang. Post-selection in noisy Gaussian boson
sampling: part is better than whole. arXiv, 2022. https://doi.org/10.48550/arXiv.2205.02586.

27
[51] L. Wright, F. Barratt, J. Dborin, G. H. Booth, and A. G. Green. Automatic post-selection by ancillae thermal-
ization. Phys. Rev. Res., 3:033151, Aug 2021.

[52] Michele Cattelan and Sheir Yarkoni. Parallel circuit implementation of variational quantum algorithms. arXiv,
2023. https://doi.org/10.48550/arXiv.2304.03037.

[53] Honghui Shang, Yi Fan, Li Shen, Chu Guo, Jie Liu, Xiaohui Duan, Fang Li, and Zhenyu Li. Towards prac-
tical and massively parallel quantum computing emulation for quantum chemistry. npj Quantum Information,
9(1):33, Apr 2023.

[54] Lana Mineh and Ashley Montanaro. Accelerating the variational quantum eigensolver using parallelism. Quan-
tum Science and Technology, 8(3):035012, may 2023.

[55] Yun-Fei Niu, Shuo Zhang, Chen Ding, Wan-Su Bao, and He-Liang Huang. Parameter-parallel distributed
variational quantum algorithm. SciPost Phys., 14:132, 2023.

[56] Keisuke Fujii, Kaoru Mizuta, Hiroshi Ueda, Kosuke Mitarai, Wataru Mizukami, and Yuya O. Nakagawa. Deep
variational quantum eigensolver: A divide-and-conquer method for solving a larger problem with smaller size
quantum computers. PRX Quantum, 3:010346, Mar 2022.

[57] Mohsin Iqbal, Nathanan Tantivasadakarn, Ruben Verresen, Sara L. Campbell, Joan M. Dreiling, Caroline
Figgatt, John P. Gaebler, Jacob Johansen, Michael Mills, Steven A. Moses, Juan M. Pino, Anthony Rans-
ford, Mary Rowe, Peter Siegfried, Russell P. Stutz, Michael Foss-Feig, Ashvin Vishwanath, and Henrik
Dreyer. Creation of non-abelian topological order and anyons on a trapped-ion processor. arXiv, 2023.
https://doi.org/10.48550/arXiv.2305.03766.

[58] Zijun Chen, Kevin J. Satzinger, Juan Atalaya, Alexander N. Korotkov, Andrew Dunsworth, Daniel Sank, Chris
Quintana, Matt McEwen, Rami Barends, Paul V. Klimov, Sabrina Hong, Cody Jones, Andre Petukhov, Dvir
Kafri, Sean Demura, Brian Burkett, Craig Gidney, Austin G. Fowler, Alexandru Paler, Harald Putterman, Igor
Aleiner, Frank Arute, Kunal Arya, Ryan Babbush, Joseph C. Bardin, Andreas Bengtsson, Alexandre Bourassa,
Michael Broughton, Bob B. Buckley, David A. Buell, Nicholas Bushnell, Benjamin Chiaro, Roberto Collins,
William Courtney, Alan R. Derk, Daniel Eppens, Catherine Erickson, Edward Farhi, Brooks Foxen, Marissa
Giustina, Ami Greene, Jonathan A. Gross, Matthew P. Harrigan, Sean D. Harrington, Jeremy Hilton, Alan
Ho, Trent Huang, William J. Huggins, L. B. Ioffe, Sergei V. Isakov, Evan Jeffrey, Zhang Jiang, Kostyantyn
Kechedzhi, Seon Kim, Alexei Kitaev, Fedor Kostritsa, David Landhuis, Pavel Laptev, Erik Lucero, Orion
Martin, Jarrod R. McClean, Trevor McCourt, Xiao Mi, Kevin C. Miao, Masoud Mohseni, Shirin Montaz-
eri, Wojciech Mruczkiewicz, Josh Mutus, Ofer Naaman, Matthew Neeley, Charles Neill, Michael Newman,
Murphy Yuezhen Niu, Thomas E. O’Brien, Alex Opremcak, Eric Ostby, Bálint Pató, Nicholas Redd, Pedram
Roushan, Nicholas C. Rubin, Vladimir Shvarts, Doug Strain, Marco Szalay, Matthew D. Trevithick, Benjamin
Villalonga, Theodore White, Z. Jamie Yao, Ping Yeh, Juhwan Yoo, Adam Zalcman, Hartmut Neven, Sergio
Boixo, Vadim Smelyanskiy, Yu Chen, Anthony Megrant, Julian Kelly, and Google Quantum AI. Exponential
suppression of bit or phase errors with cyclic error correction. Nature, 595(7867):383–387, Jul 2021.

[59] Dolev Bluvstein, Simon J. Evered, Alexandra A. Geim, Sophie H. Li, Hengyun Zhou, Tom Manovitz, Sepehr
Ebadi, Madelyn Cain, Marcin Kalinowski, Dominik Hangleiter, J. Pablo Bonilla Ataides, Nishad Maskara, Iris
Cong, Xun Gao, Pedro Sales Rodriguez, Thomas Karolyshyn, Giulia Semeghini, Michael J. Gullans, Markus
Greiner, Vladan Vuletić, and Mikhail D. Lukin. Logical quantum processor based on reconfigurable atom
arrays. Nature, December 2023.

[60] D. Coppersmith. An approximate fourier transform useful in quantum factoring, 2002. https://doi.org/
10.48550/arXiv.quant-ph/0201067.

[61] Andrew M. Childs, Richard Cleve, Enrico Deotto, Edward Farhi, Sam Gutmann, and Daniel A. Spielman. Ex-
ponential algorithmic speedup by a quantum walk. In Proceedings of the Thirty-Fifth Annual ACM Symposium

28
 on Theory of Computing, STOC ’03, page 59–68, New York, NY, USA, 2003. Association for Computing
Machinery.

[62] Gilles Brassard, Peter Høyer, Michele Mosca, and Alain Tapp. Quantum amplitude amplification and estima-
tion, pages 53–74. American Mathematical Society, 2002.

[63] Patrick Rebentrost, Adrian Steffens, Iman Marvian, and Seth Lloyd. Quantum singular-value decomposition of
nonsparse low-rank matrices. Phys. Rev. A, 97:012327, Jan 2018.

[64] Ashley Montanaro. Quantum-walk speedup of backtracking algorithms. Theory of Computing, 14(15):1–24,
2018.

[65] Andris Ambainis, Kaspars Balodis, Jānis Iraids, Martins Kokainis, Krišjānis Prūsis, and Jevgēnijs Vihrovs.
Quantum Speedups for Exponential-Time Dynamic Programming Algorithms, pages 1783–1793. Society for
Industrial and Applied Mathematics, 2019.

[66] A. Yu. Kitaev. Quantum measurements and the abelian stabilizer problem. arXiv, 1995. https://doi.
org/10.48550/arXiv.quant-ph/9511026.

[67] Stephen Jordan. Quantum Algorithm Zoo, April 2011. Last updated June 2022: https://
quantumalgorithmzoo.org/.

[68] Stuart Hadfield, Zhihui Wang, Bryan O’Gorman, Eleanor G. Rieffel, Davide Venturelli, and Rupak Biswas.
From the quantum approximate optimization algorithm to a quantum alternating operator ansatz. Algorithms,
12(2), 2019.

[69] Andrew Lucas. Ising formulations of many np problems. Frontiers in Physics, 2, 2014.

[70] M. Cerezo, Andrew Arrasmith, Ryan Babbush, Simon C. Benjamin, Suguru Endo, Keisuke Fujii, Jarrod R.
McClean, Kosuke Mitarai, Xiao Yuan, Lukasz Cincio, and Patrick J. Coles. Variational quantum algorithms.
Nature Reviews Physics, 3(9):625–644, Sep 2021.

[71] I. Ekeland. On the variational principle. Journal of Mathematical Analysis and Applications, 47(2):324–353,
1974.

[72] Alberto Peruzzo, Jarrod McClean, Peter Shadbolt, Man-Hong Yung, Xiao-Qi Zhou, Peter J. Love, Alán Aspuru-
Guzik, and Jeremy L. O’Brien. A variational eigenvalue solver on a photonic quantum processor. Nature
Communications, 5(1):4213, Jul 2014.

[73] P. J. J. O’Malley, R. Babbush, I. D. Kivlichan, J. Romero, J. R. McClean, R. Barends, J. Kelly, P. Roushan,

A. Tranter, N. Ding, B. Campbell, Y. Chen, Z. Chen, B. Chiaro, A. Dunsworth, A. G. Fowler, E. Jeffrey,
E. Lucero, A. Megrant, J. Y. Mutus, M. Neeley, C. Neill, C. Quintana, D. Sank, A. Vainsencher, J. Wenner,
T. C. White, P. V. Coveney, P. J. Love, H. Neven, A. Aspuru-Guzik, and J. M. Martinis. Scalable quantum
simulation of molecular energies. Phys. Rev. X, 6:031007, Jul 2016.

[74] E. F. Dumitrescu, A. J. McCaskey, G. Hagen, G. R. Jansen, T. D. Morris, T. Papenbrock, R. C. Pooser, D. J.

Dean, and P. Lougovski. Cloud quantum computing of an atomic nucleus. Phys. Rev. Lett., 120:210501, May
2018.

[75] Harper R. Grimsley, Sophia E. Economou, Edwin Barnes, and Nicholas J. Mayhall. An adaptive variational
algorithm for exact molecular simulations on a quantum computer. Nature Communications, 10(1):3007, Jul
2019.

[76] Google AI Quantum, Collaborators*†, Frank Arute, Kunal Arya, Ryan Babbush, Dave Bacon, Joseph C.
Bardin, Rami Barends, Sergio Boixo, Michael Broughton, Bob B. Buckley, David A. Buell, Brian Burkett,
Nicholas Bushnell, Yu Chen, Zijun Chen, Benjamin Chiaro, Roberto Collins, William Courtney, Sean Demura,

29
 Andrew Dunsworth, Edward Farhi, Austin Fowler, Brooks Foxen, Craig Gidney, Marissa Giustina, Rob Graff,
Steve Habegger, Matthew P. Harrigan, Alan Ho, Sabrina Hong, Trent Huang, William J. Huggins, Lev Ioffe,
Sergei V. Isakov, Evan Jeffrey, Zhang Jiang, Cody Jones, Dvir Kafri, Kostyantyn Kechedzhi, Julian Kelly, Seon
Kim, Paul V. Klimov, Alexander Korotkov, Fedor Kostritsa, David Landhuis, Pavel Laptev, Mike Lindmark,
Erik Lucero, Orion Martin, John M. Martinis, Jarrod R. McClean, Matt McEwen, Anthony Megrant, Xiao Mi,
Masoud Mohseni, Wojciech Mruczkiewicz, Josh Mutus, Ofer Naaman, Matthew Neeley, Charles Neill, Hart-
mut Neven, Murphy Yuezhen Niu, Thomas E. O’Brien, Eric Ostby, Andre Petukhov, Harald Putterman, Chris
Quintana, Pedram Roushan, Nicholas C. Rubin, Daniel Sank, Kevin J. Satzinger, Vadim Smelyanskiy, Doug
Strain, Kevin J. Sung, Marco Szalay, Tyler Y. Takeshita, Amit Vainsencher, Theodore White, Nathan Wiebe,
Z. Jamie Yao, Ping Yeh, and Adam Zalcman. Hartree-fock on a superconducting qubit quantum computer.
Science, 369(6507):1084–1089, 2020.

[77] Nathan Killoran, Thomas R. Bromley, Juan Miguel Arrazola, Maria Schuld, Nicolás Quesada, and Seth Lloyd.
Continuous-variable quantum neural networks. Phys. Rev. Res., 1:033063, Oct 2019.

[78] Kerstin Beer, Dmytro Bondarenko, Terry Farrelly, Tobias J. Osborne, Robert Salzmann, Daniel Scheiermann,
and Ramona Wolf. Training deep quantum neural networks. Nature Communications, 11(1):808, Feb 2020.

[79] Amira Abbas, David Sutter, Christa Zoufal, Aurelien Lucchi, Alessio Figalli, and Stefan Woerner. The power
of quantum neural networks. Nature Computational Science, 1(6):403–409, Jun 2021.

[80] Arthur Pesah, M. Cerezo, Samson Wang, Tyler Volkoff, Andrew T. Sornborger, and Patrick J. Coles. Absence
of barren plateaus in quantum convolutional neural networks. Physical Review. X, 11(4), 10 2021.

[81] Samuel Yen-Chi Chen, Chao-Han Huck Yang, Jun Qi, Pin-Yu Chen, Xiaoli Ma, and Hsi-Sheng Goan. Varia-
tional quantum circuits for deep reinforcement learning. IEEE Access, 8:141007–141024, 2020.

[82] Yunchao Liu, Srinivasan Arunachalam, and Kristan Temme. A rigorous and robust quantum speed-up in super-
vised machine learning. Nature Physics, 17(9):1013–1017, Sep 2021.

[83] Maximillian Zinner, Florian Dahlhausen, Philip Boehme, Jan Ehlers, Linn Bieske, and Leonard Fehring.
Quantum computing’s potential for drug discovery: Early stage industry dynamics. Drug Discovery Today,
26(7):1680–1688, 2021.

[84] Nick S. Blunt, Joan Camps, Ophelia Crawford, Róbert Izsák, Sebastian Leontica, Arjun Mirani, Alexandra E.
Moylett, Sam A. Scivier, Christoph Sünderhauf, Patrick Schopf, Jacob M. Taylor, and Nicole Holzmann. Per-
spective on the current state-of-the-art of quantum computing for drug discovery applications. Journal of
Chemical Theory and Computation, 18(12):7001–7023, Dec 2022.

[85] Y. Cao, J. Romero, and A. Aspuru-Guzik. Potential of quantum computing for drug discovery. IBM Journal of
Research and Development, 62(6):6:1–6:20, 2018.

[86] AstraZeneca. Artificial intelligence and machine learning: revolutionising drug development, 2017. https:
//www.astrazeneca.com/content/dam/az/PDF/2018/Pages%2046-49.pdf.

[87] Nick Paul Taylor. Alphabet’s quantum spinoff starts working with AstraZeneca, Sanofi
to accelerate R&D, June 2023. https://www.fiercebiotech.com/biotech/
alphabets-quantum-spinoff-starts-working-astrazeneca-sanofi-accelerate-rd.

[88] IBM. Cleveland Clinic and IBM Unveil First Quantum Computer Dedi-
cated to Healthcare Research, March 2023. https://newsroom.ibm.com/
2023-03-20-Cleveland-Clinic-and-IBM-Unveil-First-Quantum-Computer-Dedicated-to-He

[89] Cleveland Clinic. Discovery Accelerator, 2023. https://my.clevelandclinic.org/research/

computational-life-sciences/discovery-accelerator.

30
 [90] Suzhen Yuan, Xuefeng Mao, Jing Zhou, and Xiaofa Wang. Quantum image filtering in the spatial domain.
International Journal of Theoretical Physics, 56(8):2495–2511, Aug 2017.

[91] Simona Caraiman and Vasile I. Manta. Quantum image filtering in the frequency domain. Advances in Electri-
cal and Computer Engineering, 2013.

[92] Eduardo Bayro-Corrachono and Zuleima Vazquez-Flores. Image processing using the quantum quaternion
fourier transform. Mathematical Methods in the Applied Sciences, n/a(n/a).

[93] Amir Fijany and Colin P. Williams. Quantum wavelet transforms: Fast algorithms and complete circuits. In
Colin P. Williams, editor, Quantum Computing and Quantum Communications, pages 10–33, Berlin, Heidel-
berg, 1999. Springer Berlin Heidelberg.

[94] Sanjay Chakraborty, Soharab Hossain Shaikh, Amlan Chakrabarti, and Ranjan Ghosh. An image denoising
technique using quantum wavelet transform. International Journal of Theoretical Physics, 59(11):3348–3371,
Nov 2020.

[95] Siheon Park, Daniel K. Park, and June-Koo Kevin Rhee. Variational quantum approximate support vector
machine with inference transfer. Scientific Reports, 13(1):3288, Feb 2023.

[96] Sanjay Chakraborty, Sudhindu Bikash Mandal, and Soharab Hossain Shaikh. Quantum image processing:
challenges and future research issues. International Journal of Information Technology, 2018.

[97] Zhaobin Wang, Minzhe Xu, and Yaonan Zhang. Review of quantum image processing. Archives of Computa-
tional Methods in Engineering, 29(2):737–761, Mar 2022.

[98] Simone De Santis, Damiana Lazzaro, Riccardo Mengoni, and Serena Morigi. Quantum median filter for total
variation image denoising. ANNALI DELL’UNIVERSITA’ DI FERRARA, 68(2):597–620, Nov 2022.

[99] Xi-Wei Yao, Hengyan Wang, Zeyang Liao, Ming-Cheng Chen, Jian Pan, Jun Li, Kechao Zhang, Xingcheng
Lin, Zhehui Wang, Zhihuang Luo, Wenqiang Zheng, Jianzhong Li, Meisheng Zhao, Xinhua Peng, and Dieter
Suter. Quantum image processing and its application to edge detection: Theory and experiment. Phys. Rev. X,
7:031041, Sep 2017.

[100] Simona Caraiman and Vasile I. Manta. Image segmentation on a quantum computer. Quantum Information
Processing, 14(5):1693–1715, May 2015.

[101] XingAo Liu and Ri-Gui Zhou. Color image storage and retrieval using quantum mechanics. Quantum Infor-
mation Processing, 18(5):132, Mar 2019.

[102] Seth Lloyd, Masoud Mohseni, and Patrick Rebentrost. Quantum principal component analysis. Nature Physics,
10(9):631–633, Sep 2014.

[103] National Academies of Sciences & Engineering & Medicine. Quantum Science Concepts in Enhancing Sensing
and Imaging Technologies: Applications for Biology: Proceedings of a Workshop. The National Academies
Press, Washington, DC, 2021.

[104] Prashant S Emani, Jonathan Warrell, Alan Anticevic, Stefan Bekiranov, Michael Gandal, Michael J Mc-
Connell, Guillermo Sapiro, Alán Aspuru-Guzik, Justin T Baker, Matteo Bastiani, John D Murray, Stamatios N
Sotiropoulos, Jacob Taylor, Geetha Senthil, Thomas Lehner, Mark B Gerstein, and Aram W Harrow. Quantum
computing at the frontiers of biological sciences. Nat. Methods, 18(7):701–709, July 2021.

[105] Laura Marchetti, Riccardo Nifosì, Pier Luigi Martelli, Eleonora Da Pozzo, Valentina Cappello, Francesco
Banterle, Maria Letizia Trincavelli, Claudia Martini, and Massimo D’Elia. Quantum computing algorithms:
getting closer to critical problems in computational biology. Brief. Bioinform., 23(6), November 2022.

31
[106] L C Hollenberg. Fast quantum search algorithms in protein sequence comparisons: quantum bioinformatics.
Phys. Rev. E Stat. Phys. Plasmas Fluids Relat. Interdiscip. Topics, 62(5 Pt B):7532–7535, November 2000.

[107] Aritra Sarkar, Zaid Al-Ars, and Koen Bertels. QuASeR: Quantum accelerated de novo DNA sequence recon-
struction. PLoS One, 16(4):e0249850, April 2021.

[108] Demosthenes Ellinas and Peter Jarvis. Quantum simulation of phylogenetic trees, 2011.

[109] Salvador Elías Venegas-Andraca. Quantum walks: a comprehensive review. Quantum Information Processing,
11(5):1015–1106, July 2012.

[110] A S Boev, A S Rakitko, S R Usmanov, A N Kobzeva, I V Popov, V V Ilinsky, E O Kiktenko, and A K Fedorov.
Genome assembly using quantum and quantum-inspired annealing. Sci. Rep., 11(1):13183, June 2021.

[111] Chris Cade, Marten Folkertsma, Ido Niesen, and Jordi Weggemans. Quantum algorithms for community de-
tection and their empirical run-times, 2022.

[112] Christopher M Dobson. Chemical space and biology. Nature, 432(7019):824–828, December 2004.

[113] Jérôme Hert, John J Irwin, Christian Laggner, Michael J Keiser, and Brian K Shoichet. Quantifying biogenic
bias in screening libraries. Nat. Chem. Biol., 5(7):479–483, July 2009.

[114] Adrián Pérez-Salinas, Alba Cervera-Lierta, Elies Gil-Fuster, and José I. Latorre. Data re-uploading for a uni-
versal quantum classifier. Quantum, 4:226, February 2020.

[115] M. Cerezo, Guillaume Verdon, Hsin-Yuan Huang, Lukasz Cincio, and Patrick J. Coles. Challenges and oppor-
tunities in quantum machine learning. Nature Computational Science, 2(9):567–576, Sep 2022.

[116] W. K. Wootters and W. H. Zurek. A single quantum cannot be cloned. Nature, 299(5886):802–803, Oct 1982.

[117] Quynh T. Nguyen, Louis Schatzki, Paolo Braccia, Michael Ragone, Patrick J. Coles, Frederic Sauvage, Martin
Larocca, and M. Cerezo. Theory for equivariant quantum neural networks. arXiv, 2022.

[118] Hsin-Yuan Huang, Michael Broughton, Jordan Cotler, Sitan Chen, Jerry Li, Masoud Mohseni, Hartmut Neven,
Ryan Babbush, Richard Kueng, John Preskill, and Jarrod R. McClean. Quantum advantage in learning from
experiments. Science, 376(6598):1182–1186, 2022.

[119] Jarrod R. McClean, Sergio Boixo, Vadim N. Smelyanskiy, Ryan Babbush, and Hartmut Neven. Barren plateaus
in quantum neural network training landscapes. Nature Communications, 9(1):4812, Nov 2018.

[120] M. Cerezo, Akira Sone, Tyler Volkoff, Lukasz Cincio, and Patrick J. Coles. Cost function dependent barren
plateaus in shallow parametrized quantum circuits. Nature Communications, 12(1):1791, Mar 2021.

[121] Lennart Bittel and Martin Kliesch. Training variational quantum algorithms is np-hard. Phys. Rev. Lett.,
127:120502, Sep 2021.

[122] Johannes Bausch, Toby Cubitt, Charles Derby, and Joel Klassen. Mitigating errors in local fermionic encodings.
arXiv, 2020.

[123] Report of the Working Group on Educational Curriculum Needs for Quantum Information Sciences.
https://science.osti.gov/-/media/wdts/pdf/QIS-WG-Report.pdf, 2021. [Accessed 19-
12-2023].

[124] Luca M. Possati. Ethics of quantum computing: an outline. Philosophy & Technology, 36(3):48, Jul 2023.

[125] Ronald de Wolf. The potential impact of quantum computers on society. CoRR, abs/1712.05380, 2017.

32
[126] Dolev Bluvstein, Simon J. Evered, Alexandra A. Geim, Sophie H. Li, Hengyun Zhou, Tom Manovitz, Sepehr
Ebadi, Madelyn Cain, Marcin Kalinowski, Dominik Hangleiter, J. Pablo Bonilla Ataides, Nishad Maskara, Iris
Cong, Xun Gao, Pedro Sales Rodriguez, Thomas Karolyshyn, Giulia Semeghini, Michael J. Gullans, Markus
Greiner, Vladan Vuletić, and Mikhail D. Lukin. Logical quantum processor based on reconfigurable atom
arrays. Nature, Dec 2023.

33