Improving tractability of Clinical Decision Support system

Medical Decision Making, 1996

Neural networks are parallel, distributed, adaptive information-processing systems that develop their functionality in response to exposure to information. This paper is a tutorial for researchers intending to use neural nets for medical decision-making applications. It includes detailed discussion of the issues particularly relevant to medical data as well as wider issues relevant to any neural net application. The article is restricted to back-propagation learning in multilayer perceptrons, as this is the neural net model most widely used in medical applications. Key words: neural networks; medical decision making; pattern recognition; nonlinearity; error back-propagation; multi layer perceptron. (Med Decis Making 1996;16:386-398)

Sensors, 2022

This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY

Journal of Health & Medical Informatics, 2018

Objective: The purpose of this paper is to review the PubMed/MEDLINE literature for articles that discuss the use of machine learning (ML) and deep learning (DL) for clinical decision support systems (CDSSs). Materials and Methods: To identify relevant articles, we searched PubMed/MEDLINE through December 2 nd , 2017. We identified a total of 283 studies. Results: The number of ML and DL associated CDSS articles increased significantly beginning around 2010. The most common type of advanced artificial intelligence (AI) methodologies that the articles evaluated was neural networks also known as DL (n=109) followed by ML (n=86). The most common types of ML algorithm were support vector machines (n=78), logistic regression analysis (n=38), random forest (n=26), decision tree (n=25), and k-nearest neighbour (n=21). Cardiology, oncology, radiology, surgery, and critical care/ED were the most commonly represented specialties. Only 19 out of 283 (6.7%) ML and DL associated CDSS articles reported an effect on the process of care or patient outcomes. Discussion: The current decade has seen research efforts and attention increase significantly in creating CDSS tools with the advanced AI methodologies of DL and ML. Although the research experiments demonstrate success, the scope of AI technology is still limited to a well-defined task. Also, most of these studies lack patient-oriented outcomes necessary to justify its widespread application in healthcare. Conclusion: There is a clear upwards trend in ML and DL research in healthcare. However, in order to effectively translate successful AI research into the patient care, more clinically-relevant studies must be pursued.

This chapter provides an overview of the Machine Learning (ML) concepts in the clinical field which data may be collected, either by Health Care Professionals (HCP) or patients. These data may include activities and medication reminders, objective measurement of physiological parameters, feedback based on observed patterns, questionnaires and scores that require computational processes that give rise to useful information capable of supporting clinical decision making. The chapter describes ML in terms of learning concepts emphasizing the following approaches: supervised, unsupervised, semi-supervised, and reinforcement learning. The principles of concept classification are explained and the mathematical concepts of several methodologies are presented, such as neural networks and support vector machine among other techniques. Finally, a case study based on a radial basis function neural network aiming at the estimation of ECG waveform is presented. The proposed method reveals its suitability to support HCP on clinical decisions and practices.

Yearbook of Medical Informatics

Objectives: This survey analyses the latest literature contributions to clinical decision support systems (DSSs) on a two-year period (2017-2018), focusing on the approaches that adopt Artificial Intelligence (AI) techniques in a broad sense. The goal is to analyse the distribution of data-driven AI approaches with respect to “classical" knowledge-based ones, and to consider the issues raised and their possible solutions. Methods: We included PubMed and Web of ScienceTM publications, focusing on contributions describing clinical DSSs that adopted one or more AI methodologies. Results: We selected 75 papers, 49 of which describe approaches in the data-driven AI area, 20 present purely knowledge-based DSSs, and 6 adopt hybrid approaches relying on both formalized knowledge and data. Conclusions: Recent studies in the clinical DSS area demonstrate a prevalence of data-driven AI, which can be adopted autonomously in purely data-driven systems, or in cooperation with domain knowledg...

2012

Background Nowadays, clinical decision making is increasingly based on a large amount of patient medical data, on continuously growing medical knowledge, and on extended best clinical practice guidelines. Clinical decision support There is evidence that clinical decision support systems can significantly improve quality of care in, eventually, all areas of clinical medicine [1]. Technically, suitable means to formally represent clinical knowledge and to connect decision support algorithms with patient data sources in a seamless way are prerequisites for successful clinical decision support applications. Clinical decision support server and Arden Syntax Arden Syntax, as an internationally standardized formal language for medical knowledge representation and processing [2– 4], was implemented as a clinical decision support server and equipped with service-oriented interoperability [5]. This technical solution has already been proven to be deployable in connection with hospital and int...

Veterinary Pathology, 2019

Machine-learning methods can assist with the medical decision-making processes at the both the clinical and diagnostic levels. In this article, we first review historical milestones and specific applications of computer-based medical decision support tools in both veterinary and human medicine. Next, we take a mechanistic look at 3 archetypal learning algorithms—naive Bayes, decision trees, and neural network—commonly used to power these medical decision support tools. Last, we focus our discussion on the data sets used to train these algorithms and examine methods for validation, data representation, transformation, and feature selection. From this review, the reader should gain some appreciation for how these decision support tools have and can be used in medicine along with insight on their inner workings.

Medical Physics, 2007

We present a technique that enhances computer-assisted decision ͑CAD͒ systems with the ability to assess the reliability of each individual decision they make. Reliability assessment is achieved by measuring the accuracy of a CAD system with known cases similar to the one in question. The proposed technique analyzes the feature space neighborhood of the query case to dynamically select an input-dependent set of known cases relevant to the query. This set is used to assess the local ͑query-specific͒ accuracy of the CAD system. The estimated local accuracy is utilized as a reliability measure of the CAD response to the query case. The underlying hypothesis of the study is that CAD decisions with higher reliability are more accurate. The above hypothesis was tested using a mammographic database of 1337 regions of interest ͑ROIs͒ with biopsy-proven ground truth ͑681 with masses, 656 with normal parenchyma͒. Three types of decision models, ͑i͒ a back-propagation neural network ͑BPNN͒, ͑ii͒ a generalized regression neural network ͑GRNN͒, and ͑iii͒ a support vector machine ͑SVM͒, were developed to detect masses based on eight morphological features automatically extracted from each ROI. The performance of all decision models was evaluated using the Receiver Operating Characteristic ͑ROC͒ analysis. The study showed that the proposed reliability measure is a strong predictor of the CAD system's case-specific accuracy. Specifically, the ROC area index for CAD predictions with high reliability was significantly better than for those with low reliability values. This result was consistent across all decision models investigated in the study. The proposed case-specific reliability analysis technique could be used to alert the CAD user when an opinion that is unlikely to be reliable is offered. The technique can be easily deployed in the clinical environment because it is applicable with a wide range of classifiers regardless of their structure and it requires neither additional training nor building multiple decision models to assess the case-specific CAD accuracy.

2019

Clinical Decision Support Systems (CDSS) are computational models designed impact clinical decision making about individual patients at the point in time that these decision are made. Clinical Decision Support Systems (CDSS) form an important area of research. While traditional systematic literature surveys focus on analyzing literature using arbitrary results, visual surveys allow for the analysis of domains by using complex network-based analytical models. In this paper, we present a detailed visual survey of CDSS literature using important papers selected. The aim of this study is to review a number of articles related to CDSS for heart and stroke diseases. In this study several articles are comparable to the computational methods and rules used for data processing. From the analysis of several sources of literature, the computational methods and rules used in CDSS are Principle Component Analysis (PCA), Support Vector Machine (SVM), Naive Bayes data mining algorithm, Case Based...

2020

The main purpose of the work was to consider the problem of neural networks and their application, especially for data management and control in the medical industry. The software product, analyzes processing of unstructured and poorly structured medical data reliability, to support decision-making, implements the neural network, was developed and studied from sets of user-defined information flows. On the basis of the scientific task, the program training algorithm was developed, which provides comprehensive support for decision-making based on the study. The developed software application is focused on cross-platform, and the graphical interface is implemented using Java FX. The software product provides a network for the reverse propagation of neural network errors (BackPropagation) and a network of directed random search (Directed Random Search). Designed neural network is trained and further recognizes the type of distribution (uniform, normal) on the specified characteristics,...