Conference Presentations by Farzad V Farahani
arXiv preprint, 2020
The increasing use of deep neural networks (DNNs) has motivated a parallel endeavor: the design o... more The increasing use of deep neural networks (DNNs) has motivated a parallel endeavor: the design of adversaries that profit from successful misclassifications. However, not all adversarial examples are crafted for malicious purposes. For example, real world systems often contain physical, temporal, and sampling variability across instrumentation. Adversarial examples in the wild may inadvertently prove deleterious for accurate predictive modeling. Conversely, naturally occurring covariance of image features may serve didactic purposes. Here, we studied the stability of deep learning representations for neuroimaging classification across didactic and adversarial conditions characteristic of MRI acquisition variability. We show that representational similarity and performance vary according to the frequency of adversarial examples in the input space.
—Early detection of cancer is the most promising way to enhance a patient's chance for survival. ... more —Early detection of cancer is the most promising way to enhance a patient's chance for survival. This paper presents a computer-aided classification method using computed tomography (CT) images of the lung based on ensemble of three classifiers including MLP, KNN and SVM. In this study, the entire lung is first segmented from the CT images and specific features like Roundness, Circularity, Compactness, Ellipticity, and Eccentricity are calculated from the segmented images. These morphological features are used for classification process in a way that each classifier makes its own decision. Finally, majority voting method is used to combine decisions of this ensemble system. The performance of this system is evaluated using 60 CT scans collected by Lung Image Database Consortium (LIDC) and the results show good improvement in diagnosing of pulmonary nodules.
—Lung cancer is the second most common cancer in both men and women in the world. The focus of th... more —Lung cancer is the second most common cancer in both men and women in the world. The focus of this paper is to design a fuzzy rule based medical expert system for diagnosis of lung cancer. The proposed system consists of four modules: working memory, knowledge base, inference engine and user interface. The system takes the risk factors and symptoms of lung cancer in a two-step process and stores them as facts of the problem in working memory. Also domain expert knowledge is gathered to generate rules and stored in the rule base. The rule base consists of two different rule sets related to risk factors and symptoms of lung cancer respectively. Finally, type-2 fuzzy inference engine fires relevant rules under appropriate condition and provides the probability of disease as output of the system. The output of the system could act as a second opinion to assist the physicians. Also graphical user interface is presented to facilitate the communication between user and system.
—Lung cancer is the second most common cancer in both men and women in the world. The focus of th... more —Lung cancer is the second most common cancer in both men and women in the world. The focus of this paper is to design a fuzzy rule based medical expert system for diagnosis of lung cancer. The proposed system consists of four modules: working memory, knowledge base, inference engine and user interface. The system takes the risk factors and symptoms of lung cancer in a two-step process and stores them as facts of the problem in working memory. Also domain expert knowledge is gathered to generate rules and stored in the rule base. The rule base consists of two different rule sets related to risk factors and symptoms of lung cancer respectively. Finally, type-2 fuzzy inference engine fires relevant rules under appropriate condition and provides the probability of disease as output of the system. The output of the system could act as a second opinion to assist the physicians. Also graphical user interface is presented to facilitate the communication between user and system.
—Early detection of cancer is the most promising way to enhance a patient's chance for survival. ... more —Early detection of cancer is the most promising way to enhance a patient's chance for survival. This paper presents a computer-aided classification method using computed tomography (CT) images of the lung based on ensemble of three classifiers including MLP, KNN and SVM. In this study, the entire lung is first segmented from the CT images and specific features like Roundness, Circularity, Compactness, Ellipticity, and Eccentricity are calculated from the segmented images. These morphological features are used for classification process in a way that each classifier makes its own decision. Finally, majority voting method is used to combine decisions of this ensemble system. The performance of this system is evaluated using 60 CT scans collected by Lung Image Database Consortium (LIDC) and the results show good improvement in diagnosing of pulmonary nodules.
Papers by Farzad V Farahani
Advances in Neuroergonomics and Cognitive Engineering, 2018
Brain connectivity investigation using fMRI time series have begun since the mid-1990s and provid... more Brain connectivity investigation using fMRI time series have begun since the mid-1990s and provided a new world for researchers, especially neuroscientists, to survey the human brain network with high precision. The present study seeks to provide an overview of the computational methods available for brain connectivity, which are divided into two general categories: functional connectivity and effective connectivity. The former examines the temporal correlation between spatially remote brain areas, and the latter is about the effects of brain regions on each other. Based on these two categories of connectivity, the computational methods presented in the literature along with their strengths and weaknesses are discussed.
Brain Sciences
Task fMRI provides an opportunity to analyze the working mechanisms of the human brain during spe... more Task fMRI provides an opportunity to analyze the working mechanisms of the human brain during specific experimental paradigms. Deep learning models have increasingly been applied for decoding and encoding purposes study to representations in task fMRI data. More recently, graph neural networks, or neural networks models designed to leverage the properties of graph representations, have recently shown promise in task fMRI decoding studies. Here, we propose an end-to-end graph convolutional network (GCN) framework with three convolutional layers to classify task fMRI data from the Human Connectome Project dataset. We compared the predictive performance of our GCN model across four of the most widely used node embedding algorithms—NetMF, RandNE, Node2Vec, and Walklets—to automatically extract the structural properties of the nodes in the functional graph. The empirical results indicated that our GCN framework accurately predicted individual differences (0.978 and 0.976) with the NetMF ...
The Journal of Defense Modeling and Simulation: Applications, Methodology, Technology
Researchers and software users benefit from the rapid growth of artificial intelligence (AI) to a... more Researchers and software users benefit from the rapid growth of artificial intelligence (AI) to an unprecedented extent in various domains where automated intelligent action is required. However, as they continue to engage with AI, they also begin to understand the limitations and risks associated with ceding control and decision-making to not always transparent artificial computer agents. Understanding of “what is happening in the black box” becomes feasible with explainable AI (XAI) methods designed to mitigate these risks and introduce trust into human-AI interactions. Our study reviews the essential capabilities, limitations, and desiderata of XAI tools developed over recent years and reviews the history of XAI and AI in education (AIED). We present different approaches to AI and XAI from the viewpoint of researchers focused on AIED in comparison with researchers focused on AI and machine learning (ML). We conclude that both groups of interest desire increased efforts to obtain ...
Biology, 2022
Coronavirus disease 2019 (COVID-19) was first discovered in China; within several months, it spre... more Coronavirus disease 2019 (COVID-19) was first discovered in China; within several months, it spread worldwide and became a pandemic. Although the virus has spread throughout the globe, its effects have differed. The pandemic diffusion network dynamics (PDND) approach was proposed to better understand the spreading behavior of COVID-19 in the US and Japan. We used daily confirmed cases of COVID-19 from 5 January 2020 to 31 July 2021, for all states (prefectures) of the US and Japan. By applying the pandemic diffusion network dynamics (PDND) approach to COVID-19 time series data, we developed diffusion graphs for the US and Japan. In these graphs, nodes represent states and prefectures (regions), and edges represent connections between regions based on the synchrony of COVID-19 time series data. To compare the pandemic spreading dynamics in the US and Japan, we used graph theory metrics, which targeted the characterization of COVID-19 bedhavior that could not be explained through line...
ArXiv, 2020
The increasing use of deep neural networks (DNNs) has motivated a parallel endeavor: the design o... more The increasing use of deep neural networks (DNNs) has motivated a parallel endeavor: the design of adversaries that profit from successful misclassifications. However, not all adversarial examples are crafted for malicious purposes. For example, real world systems often contain physical, temporal, and sampling variability across instrumentation. Adversarial examples in the wild may inadvertently prove deleterious for accurate predictive modeling. Conversely, naturally occurring covariance of image features may serve didactic purposes. Here, we studied the stability of deep learning representations for neuroimaging classification across didactic and adversarial conditions characteristic of MRI acquisition variability. We show that representational similarity and performance vary according to the frequency of adversarial examples in the input space.
Frontiers in Neuroscience, 2019
Sleep is a complex and dynamic process for maintaining homeostasis, and a lack of sleep can disru... more Sleep is a complex and dynamic process for maintaining homeostasis, and a lack of sleep can disrupt whole-body functioning. No organ is as vulnerable to the loss of sleep as the brain. Accordingly, we examined a set of task-based functional magnetic resonance imaging (fMRI) data by using graph theory to assess brain topological changes in subjects in a state of chronic sleep restriction, and then identified diurnal variability in the graph-theoretic measures. Task-based fMRI data were collected in a 1.5T MR scanner from the same participants on two days: after a week of fully restorative sleep and after a week with 35% sleep curtailment. Each day included four scanning sessions throughout the day (at approximately 10:00 AM, 2:00 PM, 6:00 PM, and 10:00 PM). A modified spatial cueing task was applied to evaluate sustained attention. After sleep restriction, the characteristic path length significantly increased at all measurement times, and small-worldness significantly decreased. Assortativity, a measure of network fault tolerance, diminished over the course of the day in both conditions. Local graph measures were altered primarily across the limbic system (particularly in the hippocampus, parahippocampal gyrus, and amygdala), default mode network, and visual network.
Frontiers in Neuroscience, 2019
Conclusions: This review provides an insight into how to utilize graph theoretical measures to ma... more Conclusions: This review provides an insight into how to utilize graph theoretical measures to make neurobiological inferences regarding the mechanisms underlying human cognition and behavior as well as different brain disorders.
Mathematics and Computers in Simulation, 2018
Lung cancer is one of the most common forms of cancer leading to over a million deaths per year t... more Lung cancer is one of the most common forms of cancer leading to over a million deaths per year throughout the world. The aim of this paper is to identify the pulmonary nodules in computed tomography (CT) images of the lung using a hybrid intelligent approach. At first, the proposed approach utilizes a type-II fuzzy algorithm to improve the quality of raw CT images. Then, a novel segmentation algorithm based on fuzzy c-means clustering, called modified spatial kernelized fuzzy c-means (MSFCM) clustering, is offered in order to achieve another representation of lung regions through an optimization methodology. Next, nodule candidates are detected among all available objects in the lung regions by a morphological procedure. This is followed by extracting significant statistical and morphological features from such nodule candidates and finally, an ensemble of three classifiers comprising Multilayer Perceptron (MLP), K-Nearest Neighbor (KNN), and Support Vector Machine (SVM) is employed for the actual diagnosis and determining whether the nodule candidate is nodule (cancerous) or non-nodule (healthy). The effectiveness of the hybrid intelligent approach is evaluated using a public data set for lung CT images, viz.: Lung Image Database Consortium (LIDC). The experimental results positively demonstrate that the modified spatial kernelized FCM segmentation is superior to the other techniques existing in the literature. More importantly, a number of useful performance measurements in medical applications including accuracy, sensitivity, specificity, confusion matrix, as well as the area under the Receiver Operating Characteristic (ROC) curve are computed. The obtained results confirm the promising performance of the proposed hybrid approach in undertaking pulmonary nodules diagnosis.
The International Journal of Advanced Manufacturing Technology, 2017
The objective of the current research work was to explore the potential of emulgel in enhancing t... more The objective of the current research work was to explore the potential of emulgel in enhancing the topical delivery of piroxicam. Emulgel formulations of piroxicam were prepared using three types of gelling agents: Carbopol 934, Xanthan gum and HPMCK15M. Based on solubility studies Tween-80 and Span-80 as emulsifiers and propylene glycol as co-surfactant were selected for preparation of emulgel. The persuade of the type of the gelling agent on the drug release from the prepared emulgel was investigated. The mentha oil is used as permeation enhancer. The formulated emulgel were characterized for their physical appearance, pH determination, viscosity, spreadability, in vitro drug release, ex vivo drug release, skin irritation test, anti inflammatory activity, analgesic activity and stability studies. All the prepared formulations showed acceptable physical properties, homogeneity, consistency, spreadability, viscosity and pH value. In-vitro release study demonstrated diffusion controlled release of piroxicam from formulation up to 8 h. The drug release profile exhibited zero order kinetics.
2015 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2015
Early detection of cancer is the most promising way to enhance a patient's chance for surviva... more Early detection of cancer is the most promising way to enhance a patient's chance for survival. This paper presents a computer-aided classification method using computed tomography (CT) images of the lung based on ensemble of three classifiers including MLP, KNN and SVM. In this study, the entire lung is first segmented from the CT images and specific features like Roundness, Circularity, Compactness, Ellipticity, and Eccentricity are calculated from the segmented images. These morphological features are used for classification process in a way that each classifier makes its own decision. Finally, majority voting method is used to combine decisions of this ensemble system. The performance of this system is evaluated using 60 CT scans collected by Lung Image Database Consortium (LIDC) and the results show good improvement in diagnosing of pulmonary nodules.
2014 IEEE Conference on Norbert Wiener in the 21st Century (21CW), 2014
Multiple Sclerosis (MS) is an autoimmune disease in which insulating covers of nerves called myel... more Multiple Sclerosis (MS) is an autoimmune disease in which insulating covers of nerves called myelin sheath are damaged. Myelin sheath helps the transmission of the nerve impulses. Damage to the myelin in the central nervous system (CNS) disrupts the communication between the brain and spinal cord and other parts of the body, thus cause a wide range of signs and symptoms. Therefore, it can be difficult to diagnose by physicians in some cases. Recently automated systems have been introduced for the diagnosis of some of the neurological disorders including Multiple Sclerosis. An important issue that should be considered in these automated systems is the fact that diagnosis process often confront with uncertainty and vagueness. Therefore, we determine to bring these uncertainties in our system by using Fuzzy Logic, for first time. Another weakness seemed in previous works, is their knowledge bases and reasoning process. This paper presents a fuzzy rule-based expert system for MS diagnosis. Decision making in this system is performed based on the person's identity, symptoms and signs. In study of the cases mentioned, we confront with crisp variables that receive binary value. These crisp variables can lead to uncertain results. Fuzzy reasoning is used to address the uncertainties exist in diagnosis process. This system can help to non-neurologists in the diagnosis of MS or can be used as a neurologist physician assistant. The proposed system uses a spreadsheet for storing or extracting the information of the patients. System's knowledge base built based on direct approach and the inference is done using forward-chaining method because of the multiplicity of factors that refers to MS.
Brain Sciences, 2021
Background: Cataracts are associated with progressive blindness, and despite the decline in preva... more Background: Cataracts are associated with progressive blindness, and despite the decline in prevalence in recent years, it remains a major global health problem. Cataract extraction is reported to influence not only perception, attention and memory but also daytime sleepiness, ability to experience pleasure and positive and negative affect. However, when it comes to the latter, the magnitude and prevalence of this effect still remains uncertain. The current study aims to evaluate the hemodynamic basis of daytime sleepiness, ability to experience pleasure and positive and negative affect in cataract patients after the intraocular lens (IOL) implantation. Methods: Thirty-four cataract patients underwent resting-state functional magnetic resonance imaging evaluation before and after cataract extraction and intraocular lens implantation. Both global and local graph metrics were calculated in order to investigate the hemodynamic basis of excessive sleepiness (ESS), experiencing pleasure ...
arXiv: Other Quantitative Biology, 2020
COVID-19 is now a global pandemic, and an effective vaccine may be many months away. Over 100 yea... more COVID-19 is now a global pandemic, and an effective vaccine may be many months away. Over 100 years ago, Spanish flu fatalities were attenuated when doctors began treating patients with blood plasma donated by recovered (or convalesced) survivors. Passive immunity transfer via administration of convalesced blood product (CBP) appears to represent a readily available and promising avenue for mitigating mortalities, expediting recovery time, and even prophylaxis against the SARS-CoV-2 virus. Here, we review challenges to CBP efficacy, and present a graph theoretical model of transmission dynamics that identifies evolving hubs of COVID-19 cases. Importantly, this model suggests that CBP efficacy may rest on an efficient and distributed global sampling scheme as opposed to CBP pooled from local donors alone.
Brain Sciences
Electroencephalography (EEG) is a non-invasive technique used to record the brain’s evoked and in... more Electroencephalography (EEG) is a non-invasive technique used to record the brain’s evoked and induced electrical activity from the scalp. Artificial intelligence, particularly machine learning (ML) and deep learning (DL) algorithms, are increasingly being applied to EEG data for pattern analysis, group membership classification, and brain-computer interface purposes. This study aimed to systematically review recent advances in ML and DL supervised models for decoding and classifying EEG signals. Moreover, this article provides a comprehensive review of the state-of-the-art techniques used for EEG signal preprocessing and feature extraction. To this end, several academic databases were searched to explore relevant studies from the year 2000 to the present. Our results showed that the application of ML and DL in both mental workload and motor imagery tasks has received substantial attention in recent years. A total of 75% of DL studies applied convolutional neural networks with vario...
Circadian rhythms (lasting approximately 24 hours) control and entrain a variety of physiological... more Circadian rhythms (lasting approximately 24 hours) control and entrain a variety of physiological processes ranging from neural activity and hormone secretion to sleep cycles and feeding habits. Despite significant diurnal variation in human brain function, neuroscientists have rarely considered the effects of time-of-day (TOD) on their studies. Moreover, there are inter-individual discrepancies in sleep-wake patterns, diurnal preferences, and daytime alertness (known as chronotypes), which could be associated with human cognition and brain performance. In the present study, we performed graph-theory based network analysis on resting-state functional MRI (rs-fMRI) data to explore the topological differences in whole-brain functional networks between the morning and evening sessions (TOD effect), as well as between extreme morning-type and evening-type participants (chronotype effect). To that end, 62 individuals (31 extreme morningversus 31 evening-type) underwent two fMRI sessions:...
Uploads
Conference Presentations by Farzad V Farahani
Papers by Farzad V Farahani
Objectives: Thanks to the emergence of graph theoretical analysis, the main purpose of the current paper is to systematically review how brain properties can emerge through the interactions of distinct neuronal units in various cognitive and neurological applications using fMRI. Moreover, this article provides an overview of the existing functional and effective connectivity methods used to construct the brain network, along with their advantages and pitfalls.
Methods: In this systematic review, the databases Science Direct, Scopus, arXiv, Google Scholar, IEEE Xplore, PsycINFO, PubMed, and SpringerLink are employed for exploring the evolution of computational methods in human brain connectivity from 1990 to the present, focusing on graph theory. The Cochrane Collaboration's tool was used to assess the risk of bias in individual studies.
Results: Our results show that graph theory and its implications in cognitive neuroscience have attracted the attention of researchers since 2009 (as the Human Connectome Project launched), because of their prominent capability in characterizing the behavior of complex brain systems. Although graph theoretical approach can be generally applied to either functional or effective connectivity patterns during rest or task performance, to date, most articles have focused on the resting-state functional connectivity.
Conclusions: This review provides an insight into how to utilize graph theoretical measures to make neurobiological inferences regarding the mechanisms underlying human cognition and behavior as well as different brain disorders.