Digital Commons Network^™

Highly Effective Combination Of Brg1/Brm Inhibitor With Bet Inhibitor Or Decitabine For High-Risk Mecom-Rearranged Aml, Warren Fiskus, Christopher P Mill, Jessica Piel, Mike Collins, Murphy Hentemann, Branko Cuglievan, Christine E Birdwell, Kaberi Das, John A Davis, Hanxi Hou, Antrix Jain, Anna Malovannaya, Lauren B Flores, Tapan M Kadia, Naval Daver, Koji Sasaki, Koichi Takahashi, Danielle Hammond, Jian Wang, Sanam Loghavi, Xiaoping Su, Courtney D Dinardo, Ruud Delwel, Kapil N Bhalla

Faculty, Staff and Students Publications

In AML with 3q26.2 rearrangements (r) the distal GATA2 hematopoietic enhancer becomes aberrantly relocated leading to activation of EVI1 expression. EVI1 is a transcriptional regulator that plays a role in proliferation and maintenance of a stem cell-like phenotype in AML. BRG1 (SMARCA4) and BRM (SMARCA2) are the mutually exclusive ATPases of the BAF (BRG1/BRM-associated factor) chromatin remodeling complexes. They regulate access to enhancers/promoters and gene-expressions orchestrating AML stem/progenitor cell proliferation and differentiation. AML with 3q26.2 rearrangements are clinically challenging and prognosis remains very poor. FHD-286 is an orally bioavailable, selective inhibitor of BRG1/BRM under clinical development in AML. Present studies …

Gerd Mastery For The Advanced Practice Provider: A Train-The Trainer Program, Gwen M. Cassidy

Yale School of Nursing Digital Theses

AbstractGERD Mastery for the Advanced Practice Provider: A Train-The Trainer Program

Background: This Doctor of Nursing Practice (DNP) project addressed a specialty-care gap in gastroenterology driven by the projected U.S. shortage of ~1,630 gastroenterology physicians. Gastroesophageal reflux disease (GERD) has a high prevalence and cost burden as is, but delayed treatment can further increase costs (Barrett’s esophagus ~$8,755/year; esophageal adenocarcinoma ~$146,391/year) and patient morbidity.

Purpose: To develop and implement “GERD Mastery for the Advanced Practice Provider,” an evidence-based train-the-trainer (TTT) program to increase guideline-concordant GERD knowledge and train others. Methods: In partnership with a national organization, a Plan–Do–Study–Act framework guided …

A Model For Virtually Integrated Care In A Long-Term Care Hospital: Improving Nurse Satisfaction While Reducing Overtime Costs, Lisa Kalafus

Yale School of Nursing Digital Theses

AbstractA Model for Virtually Integrated Care in a Long-Term Care Hospital: Improving Nurse Satisfaction while Reducing Overtime Costs Challenges related to workforce shortages, along with recruitment and retention are projected to continue for years to come as the need for health care grows and the demand for nurses exceeds the supply. Hospitals must continue to focus on the reduction of nurse turnover and proactively address underlying causes of nurse dissatisfaction. Increased workload has a negative impact on job satisfaction. The documentation requirements alone are considered a top indicator of nurse workload and take a significant amount of time away from …

Mental Health Screening And Referral For Adults Experiencing Homelessness, Jennifer Lee Ramsey

Yale School of Nursing Digital Theses

Mental health disorders affect 26–50% of the 541,484 adults experiencing homelessness in the United States, with a bidirectional relationship between homelessness and mental illness that perpetuate each other, reinforcing difficulties maintaining housing and employment, straining relationships, worsening psychiatric symptoms, and increasing barriers to care. (Nilsson et al., 2024; U.S. Department of Housing and Urban Development, 2023; Fazel et al., 2014; National Coalition for the Homeless, 2017; Saldua, 2024).This Doctor of Nursing Practice (DNP) project addresses these disparities by developing and implementing a mental health screening and referral program at a homeless shelter. The program used the Mood Disorder Questionnaire (MDQ), …

Multi-Omics To Study Chronic Respiratory Diseases And Viral Infections, Sobia Idrees, Hao Chen, Tayyaba Sadaf, Saima Firdous Rehman, Matt D Johansen, Keshav Raj Paudel, Gang Liu, Yuting Wang, Malte D Luecken, Elinor Hortle, Ashleigh S Philp, Kurtis F Budden, Matthew O'Rourke, Gerard E Kaiko, Sionne E M Lucas, Joanne L Dickinson, Peter C Allen, Joseph E Powell, Lai-Ying Zhang, Daniel C Chambers, Tamera Corte, Gaetano Caramori, Maor Sauler, Peter A Wark, Janine Gote-Schniering, Mareike Lehmann, Thomas M Conlon, Theodore S Kapellos, Ali Önder Yildirim, Rosa Faner, Shyamali C Dharmage, Craig E Wheelock, Maarten Van Den Berge, Martijn C Nawijn, Francesca Polverino, Gabrielle T Belz, Sanjay H Chotirmall, Leopoldo N Segal, Alen Faiz, Philip M Hansbro

Faculty, Staff and Students Publications

Integrating omics layers reveals complex biological systems, unveiling vital insights into chronic respiratory diseases and COVID-19. This holistic approach is crucial for developing more effective treatments and understanding intricate relationships. https://bit.ly/3VKFzJH

Cryogenic Sloshing In Aircraft Fuel Tanks, Stuart Colville

School of Engineering, Computing and Mathematics Theses

Liquid sloshing in next-generation sub-cooled liquid hydrogen aircraft fuel tanks can induce rapid ullage pressure and temperature drops, potentially causing cavitation in cryogenic pumping systems and compromising fuel delivery systems. Sloshing events may be initiated during taxiing, take-off, landing, and turbulence, with large accelerations producing highly non-linear liquid motion and wave-breaking conditions. Understanding such phenomena is essential for cryogenic hydrogen fuel tank design and certification for next generation aircraft.A systematic experimental methodology was employed to investigate both the wave kinematic and thermodynamic aspects of sloshing. Initial water–air experiments in a simplified horizontal circular tank were conducted to validate measurement techniques, …

Reframing Trauma-Informed Care For Refugee And (Im)Migrant Health Systems: From Interpersonal Practice To Structurally Embedded Care, Hyojin Im, Sophia Banu, Aisha Koroma, Sarmaya Mustafayeva

Faculty, Staff and Students Publications

No abstract provided.

Using Lotilaner Ophthalmic Solution, 0.25% To Treat Demodex Blepharitis In Patients Who Were Initially Misdiagnosed: A Case Series, Laura M Periman, Richard Lindstrom, Paul Karpecki, Brandon Ayres, Eric Donnenfeld, Marjan Farid, Ian Benjamin Gaddie, Preeya K Gupta, Selina Mcgee, Kelly K Nichols, Stephen Pflugfelder, Christopher E Starr, Nandini Venkateswaran, Elizabeth Yeu, Cecelia Koetting

Faculty, Staff and Students Publications

Introduction: Demodex blepharitis (DB) is a chronic ocular inflammation caused by Demodex mite infestation of the eyelid. DB is often misdiagnosed, as some of its clinical findings and patient-reported outcomes overlap with those of other ocular surface diseases. The Demodex Expert Panel on Treatment and Eyelid Health (DEPTH) has reached consensus on the symptomatology, diagnosis, and treatment of DB. Consensus findings from the DEPTH group include that collarettes are pathognomonic for DB and that lotilaner ophthalmic solution, 0.25%, the first and only FDA-approved treatment for DB, should be the first-line treatment for the disease. This case series highlights two instances …

Non-Congenital Hsv-1 Keratitis In An Immunocompetent 9-Month-Old Infant, Caleb A Casanova, Stephanie Trejo Corona, Stephen F Pflugfelder

Faculty, Staff and Students Publications

Isolated, non-congenital herpes simplex virus type 1 (HSV-1) keratitis in immunocompetent infants is exceedingly rare and may be easily misdiagnosed, leading to delays in appropriate antiviral therapy. This case is unique in that it describes HSV-1 keratitis in a healthy 9-month-old infant presenting without systemic illness, initially treated as bacterial conjunctivitis. The patient presented with unilateral eye pain, redness, and corneal opacification. Subsequent evaluation revealed HSV-1 keratitis, and the patient was treated with intravenous acyclovir in combination with topical moxifloxacin and topical ganciclovir, resulting in complete resolution of symptoms within one week. This case adds to the limited literature on …

Combination Therapy With 308-Nm Excimer Laser, Compound Glycyrrhizin, And Tacrolimus For Pediatric Facial And Cervical Vitiligo, Yingying Qian, Yafen Wu, Wei Li, Qian Chen, Siyuan Song, Ting Zhang

Faculty, Staff and Students Publications

Background: Vitiligo is a common acquired depigmenting disorder in pediatric dermatology. Facial and cervical involvement is particularly distressing due to cosmetic disfigurement, leading to psychosocial impairment in children and significant psychological burden for parents. Safe and effective treatment strategies to halt disease progression and promote repigmentation are urgently needed.

Objective: To evaluate the clinical efficacy, onset of repigmentation, safety, and repigmentation patterns of 308-nm excimer laser combined with oral compound glycyrrhizin and topical 0.03% tacrolimus in children with facial and cervical vitiligo.

Methods: A total of 112 pediatric patients were randomized into two groups: treatment group (n=56, 86 lesions) received …

3d Geometric Modelling And Clinical Validation Of Lower Limb And Preliminary 2-Year Post-Operative Clinical Outcomes Of Neutral Boundary Alignment In Total Knee Arthroplasty, Bruno Violante, Lorenzo Deveza, Francesco Pollara, Giovanni Rusconi, Gianroberto Ferreri, Alessandro Annibaldi, Ilwhan Park

Faculty, Staff and Students Publications

Purpose: To introduce a novel three-dimensional (3D) modelling concept for lower limb alignment in total knee arthroplasty (TKA), termed neutral boundary alignment (NBA) and to present preliminary clinical data supporting this physics-based custom alignment strategy.

Methods: NBA is defined as the alignment at which the knee joint line is parallel to the ground during the mid-stance phase of gait, achieving static equilibrium with the NBA axis aligned to gravity. A theoretical geometric model of the lower limb was developed, integrating principles of physics to describe this condition. Preliminary clinical validation was performed using radiographic data from an initial cohort of …

Functional Analysis Of Genes Enriched In Male Reproductive Organs: Validation Via Fertility Assessment Of 14 Knockout Mouse Lines, Kazumasa Takemoto, Katarzyna Kent, Kaori Nozawa, Shingo Tonai, Ahn H Pham, Terumi Fujimoto, Thomas X Garcia, Martin M Matzuk, Masahito Ikawa, Keizo Tokuhiro

Faculty, Staff and Students Publications

Tissue-restricted gene expression serves as a key indicator of functional importance within specific organs. Despite advances in high-throughput sequencing and genome editing over the past decade, this principle remains the gold standard for identifying candidate genes for functional studies. In reproductive biology, testis-restricted expression is anticipated to indicate genes crucial for gametogenesis, given the conserved features shared between spermatogenesis and oogenesis. Concurrently, rapid evolution of spermatogenesis-related genes is well documented and thought to arise through duplication and subsequent mutation of ubiquitously expressed genes. In this study, we established 14 knockout (KO) mouse lines for male reproductive organ-enriched genes, all of …

Highly Variable Expressivity Of A Cnv Deletion Involving Tbx4 In Three Deceased Siblings With Lung Developmental Disorder And Their Mildly Affected Mother And Grandfather, Przemyslaw Szafranski, Tomasz Gambin, Michal Kadlof, Michał Denkiewicz, Dariusz Plewczynski, Hyun Jeong Kim, Gail Deutsch, Nahir Cortes-Santiago, Salmo Raskin, Paweł Stankiewicz

Faculty, Staff and Students Publications

Single nucleotide variants (SNVs) and copy-number variant (CNV) deletions involving TBX4 have been associated with pulmonary arterial hypertension, ischiocoxopodopatellar syndrome, and lethal lung developmental disorders (LLDDs). Thus far, all large CNV deletions encompassing entire TBX4 have been found to have arisen de novo. Here, we present a three-generation family with three neonate siblings who died within 35-66 days due to histopathologically diagnosed LLDD. Whole-genome sequencing identified an ~108-kb CNV deletion encompassing TBX4 in all three infants. The deletion was also found in their mother with a history of pneumonia and persistent thick upper airway secretions and in the maternal grandfather …

Noradrenergic Innervation Across Brain Regions Is Altered By Aging And By Disease Progression In A Mouse Model Of Alzheimer’S Disease Neuropathology, Nicole M Hernandez, Manuel Silva-Pérez, Jeannie Chin

Faculty, Staff and Students Publications

Norepinephrine plays critical roles in modulating arousal and attention, is highly dynamic in awake, behaving individuals, and has anti-inflammatory and neuroprotective actions. Notably, the locus coeruleus (LC), the primary source of norepinephrine in the central nervous system, is among the first brain regions to show pathological alterations in early stages of Alzheimer's disease (AD). LC neuronal loss and associated reductions in norepinephrine in the brain have therefore been postulated to play a key role in AD pathophysiology. LC neurons and their axons have been studied in several mouse models of AD-related neuropathology to investigate their contribution to brain dysfunction in …

Frontline Aspiration Versus Stent Retriever Thrombectomy For M2 Occlusions: Insights From The Star Registry, Michael Gaub, Rahim Abo Kasem, Ilko Maier, Ansaar Rai, Pascal Jabbour, Joon-Tae Kim, Brian Howard, Ali Alawieh, Stacey Quintero Wolfe, Robert M Starke, Marios-Nikos Psychogios, Amir Shaban, Nitin Goyal, Justin Dye, Ali Alaraj, Mohamad Ezzeldin, Shinichi Yoshimura, David Fiorella, Omar Tanweer, Daniele G Romano, Pedro Navia, Hugo Cuellar, Isabel Fragata, Adam Polifka, Joshua Osbun, Fazeel Siddiqui, Mark Moss, Kaustubh Limaye, Maxim Mokin, Charles Matouk, Min S Park, Waleed Brinjikji, Ergun Daglioglu, Richard Williamson, David J Altschul, Christopher S Ogilvy, Roberto Crosa, Michael R Levitt, Benjamin Gory, Alexandra Paul, Peter Kan, Walter Casagrande, Shakeel Chowdhry, Michael F Stiefel, Ramesh Grandhi, Alejandro Spiotta, Justin Mascitelli

Faculty, Staff and Students Publications

Background: Recent trials have furthered uncertainty regarding the endovascular benefit for medium vessel occlusions (MeVO). Stent retrievers (SR) were employed in the first attempt in most interventional arm participants. We sought to compare outcomes in acute MCA M2 occlusions between frontline aspiration and SR, and to delineate procedural and anatomical covariates associated with differential treatment effect.

Methods: Retrospective analysis of a multicenter stroke thrombectomy cohort identified cases of MT for M2 occlusions. Unmatched and propensity score-matched (PSM) cohorts were generated comparing frontline aspiration to standalone and combined SR. The primary outcome was functional independence (mRS 0-2) at 90 days. Recanalization, …

Comparing Rotational Thromboelastometry And Standard Coagulation Assays For Predicting Intraoperative Bleeding In Pediatric Liver Transplantation, Kirby Deshotels, Trung Nguyen, Jun Teruya, Muhammed Umair M Mian, Kelby Fuller, Sanjiv Harpavat, Anna Banc-Husu, Amir Navaei, John Goss, Moreshwar Desai, Arun Saini

Faculty, Staff and Students Publications

Background: Utility of preoperative rotational thromboelastometry (ROTEM) over standard coagulation assays (SCAs) in predicting intraoperative bleeding during orthotopic liver transplantation (OLT) in children with liver failure (LF) remains unclear.

Methods: In this single-center retrospective cohort of pediatric OLT recipients, we compared the predictive values of preoperative ROTEM parameters (intrinsic pathway (INTEM) maximum clot firmness (MCF), extrinsic pathway (EXTEM) clotting time (CT), INTEM CT, and fibrinogen-based (FIBTEM) MCF) and the corresponding SCAs (platelet count, international normalized ratio (INR), activated partial thromboplastin time (aPTT), and fibrinogen level, respectively) for significant intraoperative bleeding (blood loss of > 22 mL/kg; i.e., ≥ 85th percentile for …

Untrained Position-Encoded Multilayer Perceptron Network For Structured Illumination Microscopy Reconstruction, Sahil Sharma, Leonidas Zimianitis, Krishnendu Samanta, Balpreet Singh Ahluwalia, Joby Joseph, Dushan N. Wadduwage

Computer Science Faculty Publications

Structured Illumination Microscopy (SIM) enables super-resolution imaging by encoding high-frequency spatial information through patterned light. While traditional Fourier-based reconstruction methods are prone to artifacts under suboptimal conditions, recent deep learning approaches often require large training datasets and lack adaptability across different imaging setups. In this work, we present Position Encoded Multi-Layer Perceptron (PEM) network that leverages implicit neural representations (INRs) and SIM forward-model-driven modeling to reconstruct super-resolved images without any training data. PEM-SIM represents each spatial coordinate as a combination of sinusoidal functions across multiple frequencies, enabling rich encoding of fine spatial detail. A forward model grounded in SIM image …

Application Paths Of Semantic Modeling In Financial Fraud Detection And Risk Identification, Victor P. Gauthier, Daniel S. Wu

Computer Science Faculty Publications

Financial fraud and risk pose significant threats to economic stability and individual well-being. Traditional detection methods often struggle to keep pace with increasingly sophisticated fraudulent schemes. Semantic modeling, which focuses on understanding the meaning and relationships within data, offers a promising avenue for enhancing fraud detection and risk identification. This review paper explores the application paths of semantic modeling in this domain. We begin with a historical overview of fraud detection techniques, highlighting the limitations of traditional approaches. Subsequently, we delve into core themes, including knowledge graph-based fraud detection and semantic rule-based inference for risk assessment. We then compare and …

Integration Of Hybrid Quantum-Neuromorphic Ai With Cloud, Edge, And High-Performance Computing Environments, Arun B. Prasad, Ajay Prasad, Dineshkumar Rajendran, Anurag Tiwari, T. Akilan, Islombek Khushvaktov

Computer Science Faculty Publications

The convergence of quantum computing, neuromorphic learning, and distributed cloud infrastructures has occurred very rapidly, and intelligent systems are now providing new opportunities, but the challenge of instability, complexity of orchestration, and noise sensitivity remains in the way of practical integration. The proposed work is based on a hybrid quantum and neuromorphic architecture, which is the integration of event-based neuromorphic adaptation and quantum-assisted global optimization, orchestrated by cloud-HPC. The architecture presents the thermodynamically regularized learning and resourceful task scheduling to the probabilistic search and the continuous local adaptation. Experimental evaluation across financial modeling, medical imaging, and physical system prediction shows …

Cognitive Prosthetic: An Ai-Enabled Multimodal System For Episodic Recall In Knowledge Work, Lawrence Obiuwevwi, Krzystof J. Rechowicz, Vikas Ashok, Sachin Shetty, Sampath Jayarathna

Computer Science Faculty Publications

Modern knowledge workplaces increasingly strain human episodic memory as individuals navigate fragmented attention, overlapping meetings, and multimodal information streams. Existing workplace tools provide partial support through note-taking or analytics but rarely integrate cognitive, physiological, and attentional context into retrievable memory representations. This paper presents the Cognitive Prosthetic Multimodal System (CPMS)—an AI-enabled proof-of-concept designed to support episodic recall in knowledge work through structured episodic capture and natural language retrieval. CPMS synchronizes speech transcripts, physiological signals, and gaze behavior into temporally aligned, JSON-based episodic records processed locally for privacy. Beyond data logging, the system includes a web-based retrieval interface that allows users …

Exploring Marshall–Olkin Models Through Bibliometric And Topic Modeling Approaches Uses Latent Dirichlet Allocation (1981-2025): A Study Based On Scopus Data, Humberto Llinás, Brian Llinás, Carlos López, Daniela Nuñez

Computer Science Faculty Publications

The Marshall–Olkin family of distributions has gained increasing attention in fields such as reliability engineering, survival analysis, financial risk modeling, and actuarial science because of its flexibility in modeling dependence among events and its wide range of extensions. Despite its growing relevance, a systematic understanding of how research on Marshall–Olkin models has evolved over time is still limited. This study addresses this gap by combining bibliometric techniques with topic modeling to analyze the structure and evolution of the scientific literature on Marshall–Olkin models. The analysis includes all 266 peer-reviewed publications on Marshall–Olkin models indexed in Scopus between 1981 and 2025. …

Stochastic Fractional-Order Memristive Fuzzy Bam Neural Networks With Time Delays And Leakage Term For Finite-Time Stability Analysis, J. Kumar, M. Syed Ali, Sumaya Sanober, Mohammad Yarish, Abeer M. Alotaibi, Tarek F. Ibrahim

Computer Science Faculty Publications

In this study, a finite-time stability analysis with time delays and a leakage term is conducted on stochastic fractional-order memristive fuzzy BAM neural networks. FOMFBAMNNs are developed using set-valued map theories as well as differential inclusion. We obtained several significant adequate criteria of uniform stability in the mean square of such networks by using analytical methods and inequality approaches, such as Cauchy–Schwarz inequality and Burkholder–Davis–Gundy inequality. In addition to examining two different fractional-order derivatives between the U-layer and V-layer synchronously with fractional order, the existence, uniqueness, and stability of its equilibrium point are also shown ½ ≤ α ≤ 1. …

Towards Supporting Real-Time Estimation Of Vehicle Fuel Consumption And Co2 Emissions In Smart City Applications, Abrar Alali, Stephan Olariu

Computer Science Faculty Publications

This paper evaluates a simplified physics-based energy demand model designed to estimate vehicle fuel consumption and CO₂ emissions—a critical tool for sustainable transportation planning and smart city applications. Unlike data-driven regression models that lack generalizability for user-defined conditions or complex physics-based approaches that rely on extensive, often proprietary data, the simplified model is distinguished by its minimal parameter requirements, depending primarily on a single, overarching powertrain efficiency value. A key contribution is the comprehensive empirical evaluation of the simplified model against official Environmental Protection Agency (EPA) test data across multiple driving cycles and vehicle types, providing a rigorous validation previously …

Contextual Scaffolding And Self-Efficacy: Supporting Computer Skill Development Among Blind Learners In India, Akshay Kolgar Nayak, Yash Prakash, Sampath Jayarathna, Hae-Na Lee, Vikas Ashok

Computer Science Faculty Publications

Inclusive computer literacy education efforts, broadening the participation of blind or visually impaired (BVI) individuals, have gained traction in recent years. Existing literature investigating these efforts primarily draws evidence from affluent Global North contexts, where accessibility resources and legal frameworks are relatively more mature. Little is known about the in-situ teaching and learning challenges faced by trainers and BVI students, respectively, in resource-constrained, multicultural Global South countries like India. To address this knowledge gap, we conducted a four-month contextual inquiry at two computer training centers catering to 94 BVI students in India. We notably observed a rigid, experience-driven training environment …

Open Scholarly Information Systems: Status Quo, Challenges, Opportunities, Hannah Bast, Guillaume Cabanac, Paolo Manghi, Jian Wu, Marcel R. Ackermann

Computer Science Faculty Publications

Over the past 30 years, a rich ecosystem of scholarly information systems has developed that openly provide their services to the scientific community. These systems include aggregators of bibliographic metadata (e.g., DBLP, OpenCitations, OpenAIRE Graph, OpenAlex, ORKG, Semantic Scholar, CiteSeerX, and CORE); publication, data, and software repositories (e.g., Arxiv.org, Figshare, Zenodo, Software Heritage, and Dataverse); and PID authorities (e.g., ORCID, ROR, Crossref, and DataCite). This interdisciplinary Dagstuhl Seminar "Open Scholarly Information Systems: Status Quo, Challenges, Opportunities" (25381) was the first of its kind to bring together practitioners from this ecosystem, as well as researchers investigating related questions or relying on …

Finding The Signal In The Noise: An Exploratory Study On Assessing The Effectiveness Of Ai And Accessibility Forums For Blind Users' Support Needs, Satwik Ram Kodandaram, Jiawei Zhou, Xiaojun Bi, Iv Ramakrishnan, Vikas Ashok

Computer Science Faculty Publications

Accessibility forums and, more recently, generative AI tools have become vital resources for blind users seeking solutions to computer-interaction issues and learning about new assistive technologies, screen reader features, tutorials, and software updates. Understanding user experiences with these resources is essential for identifying and addressing persistent support gaps. Towards this, we interviewed 14 blind users who regularly engage with forums and GenAI tools. Findings revealed that forums often overwhelm users with multiple overlapping topics, redundant or irrelevant content, and fragmented responses that must be mentally pieced together, increasing cognitive load. GenAI tools, while offering more direct assistance, introduce new barriers …

Explainable Convolutional Neural Network Model Provides An Alternative Genome-Wide Association Perspective On Mutations In Sars-Cov-2, Parisa C. Hatami, Richard Annan, Luis Miranda, Jane L. Gorman, Mengjun Xie, Letu Qingge, Hong Qin

Computer Science Faculty Publications

Identifying informative genomic features in SARS-CoV-2 can help clarify patterns of viral evolution. In this study, we developed an explainable convolutional neural network (CNN) model to classify SARS-CoV-2 genomic sequences into the WHO-designated Variants of Concern (VOCs), Alpha, Beta, Gamma, Delta, and Omicron. Using a balanced dataset of genomes, the classification CNN achieved 99.96% accuracy on the held-out test set. To interpret the model’s predictions, we applied SHapley Additive exPlanations (SHAP) to estimate the contribution of each nucleotide position to VOC-label prediction and compared aggregated attributions with a chi-square GWAS baseline applied to the same categorical labels. SHAP prioritized several …

A Comparative Analysis Of Explainable Ai (Xai) Techniques For Transparent And Reliable Image Classification, Sovon Chakraborty, Shakib Mahmud Dipto, Kevin R. Pilkiewicz, Michael L. Mayo, Pratip Rana

Computer Science Faculty Publications

Evaluating the trustworthiness of black-box machine learning models remains a significant methodological challenge. Their lack of transparency and interpretability limits applicability, because stakeholders often seek transparency before trusting the results of black-box machine learning models. Explainable AI (XAI) methods provide for human-understandable justifications and informed decision-making of these black-box architectures. Therefore, it is imperative to select the proper XAI model tailored to specific tasks. In this research, we focus on examining four XAI techniques: PEEK, LRP, GRAD-CAM, and LIME to understand how they perform against each other for image classification tasks. We evaluate the performance, robustness, generalizability, noise stability, and …

Guidelines For Automatic Grading Of Student Essays Using Large Language Models, Diwakar Yalpi, Sruta Keerti Kasula, Ravi Mukkamala

Computer Science Faculty Publications

Automated essay evaluation using large language models (LLMs) has emerged as a promising approach to support scalable and consistent educational assessment. However, the effectiveness of LLM-based grading varies significantly across evaluation dimensions and is highly influenced by prompt design and model selection. In this study, we evaluate five state-of-the-art LLMs across five rubric-based categories: Relevance to Question, Reasoning and Critical Thinking, Evidence and Examples, Organization, and Clarity and Writing Quality. We systematically investigate the impact of three prompting strategies, including rubric-only prompting, exemplar-based prompting (with and without rubric guidance)(Original and Refined prompt designs) incorporating structured instructions. Additionally, a prompt ablation …

Toward An Event-Level Analysis Of Hadron Structure Using Differential Programming, Kevin Braga, Markus Diefenthaler, Steven Goldenberg, Daniel Lersch, Yaohang Li, Jian-Wei Qiu, Kishansingh Rajput, Felix Ringer, Nobuo Sato, Malachi Schram

Computer Science Faculty Publications

Reconstructing the internal properties of hadrons in terms of fundamental quark and gluon degrees of freedom is a central goal in nuclear and particle physics. This effort lies at the core of major experimental programs, such as the Jefferson Lab 12 GeV program and the upcoming Electron-Ion Collider. A primary challenge is the inherent inverse problem: converting large-scale observational data from collision events into the fundamental quantum correlation functions (QCFs) that characterize the microscopic structure of hadronic systems within the theory of QCD. Recent advances in scientific computing and machine learning have opened new avenues for addressing this challenge using …