Medicine and Health Sciences Commons^™

Differential Antibody Response To Ebv Proteome Following Ebvst Immunotherapy In Ebv-Associated Lymphomas, Yomani D Sarathkumara, Nathan W Van Bibber, Zhiwei Liu, Helen E Heslop, Rayne H Rouce, Anna E Coghill, Cliona M Rooney, Carla Proietti, Denise L Doolan

Faculty, Staff and Students Publications

Epstein-Barr virus (EBV) is associated with a diverse range of lymphomas. EBV-specific T-cell (EBVST) infusions have shown promise in safety and clinical effectiveness in treating EBV-associated lymphomas; however, not all patients respond to T-cell immunotherapies. To identify EBV antigen–specific antibody responses associated with clinical outcomes, we comprehensively characterized antibody responses to the complete EBV proteome using a custom protein microarray in 56 patients with EBV-associated lymphoma who received EBVST infusions in phase 1 clinical trials. Responders (nonprogressors) and nonresponders (progressors) had distinct antibody profiles against EBV. Twenty-five immunoglobulin G (IgG) antibodies were significantly elevated in higher levels in nonresponders than …

Mri-Based Digital Twins To Improve Treatment Response Of Breast Cancer By Optimizing Neoadjuvant Chemotherapy Regimens, Chengyue Wu, Ernesto A B F Lima, Casey E Stowers, Zhan Xu, Clinton Yam, Jong Bum Son, Jingfei Ma, Gaiane M Rauch, Thomas E Yankeelov

Faculty, Staff and Student Publications

We developed a practical framework to construct digital twins for predicting and optimizing triple-negative breast cancer (TNBC) response to neoadjuvant chemotherapy (NAC). This study employed 105 TNBC patients from the ARTEMIS trial (NCT02276443, registered on 10/21/2014) who received Adriamycin/Cytoxan (A/C)-Taxol (T). Digital twins were established by calibrating a biology-based mathematical model to patient-specific MRI data, which accurately predicted pathological complete response (pCR) with an AUC of 0.82. We then used each patient's twin to theoretically optimize outcome by identifying their optimal A/C-T schedule from 128 options. The patient-specifically optimized treatment yielded a significant improvement in pCR rate of 20.95-24.76%. Retrospective …

A Folding-Docking-Affinity Framework For Protein-Ligand Binding Affinity Prediction, Ming-Hsiu Wu, Ziqian Xie, Degui Zhi

Faculty, Staff and Student Publications

Accurate protein-ligand binding affinity prediction is crucial in drug discovery. Existing methods are predominately docking-free, without explicitly considering atom-level interaction between proteins and ligands in scenarios where crystallized protein-ligand binding conformations are unavailable. Now, with breakthroughs in deep learning AI-based protein folding and binding conformation prediction, can we improve binding affinity prediction? This study introduces a framework, Folding-Docking-Affinity (FDA), which folds proteins, determines protein-ligand binding conformations, and predicts binding affinities from three-dimensional protein-ligand binding structures. Our experimental results indicate that FDA performs comparably to state-of-the-art docking-free methods. We anticipate that our proposed framework serves as a starting point for integrating …

A Scoping Review Of Omop Cdm Adoption For Cancer Research Using Real World Data, Liwei Wang, Andrew Wen, Sunyang Fu, Xiaoyang Ruan, Ming Huang, Rui Li, Qiuhao Lu, Heather Lyu, Andrew E Williams, Hongfang Liu

Faculty, Staff and Student Publications

The Observational Medical Outcomes Partnership (OMOP) common data model (CDM) supports large-scale research by enabling distributed network analyses. However, the breadth of its adoption in cancer research is not well understood. We conducted a scoping review to describe the adoption of the OMOP CDM in cancer research. A total of 49 unique articles were included in the review, with 30 on the data analysis theme, and 20 on the infrastructure theme. This review highlighted that while the OMOP CDM ecosystem has enabled successful data support for cancer research, particularly for collaborative studies, ongoing model development and iterative improvement remain needed …

Benchmarking Large Language Models For Biomedical Natural Language Processing Applications And Recommendations, Qingyu Chen, Yan Hu, Xueqing Peng, Qianqian Xie, Qiao Jin, Aidan Gilson, Maxwell B Singer, Xuguang Ai, Po-Ting Lai, Zhizheng Wang, Vipina K Keloth, Kalpana Raja, Jimin Huang, Huan He, Fongci Lin, Jingcheng Du, Rui Zhang, W Jim Zheng, Ron A Adelman, Zhiyong Lu, Hua Xu

Faculty, Staff and Student Publications

The rapid growth of biomedical literature poses challenges for manual knowledge curation and synthesis. Biomedical Natural Language Processing (BioNLP) automates the process. While Large Language Models (LLMs) have shown promise in general domains, their effectiveness in BioNLP tasks remains unclear due to limited benchmarks and practical guidelines. We perform a systematic evaluation of four LLMs-GPT and LLaMA representatives-on 12 BioNLP benchmarks across six applications. We compare their zero-shot, few-shot, and fine-tuning performance with the traditional fine-tuning of BERT or BART models. We examine inconsistencies, missing information, hallucinations, and perform cost analysis. Here, we show that traditional fine-tuning outperforms zero- or …

Tobacco Smoke Exposure Is A Driver Of Altered Oxidative Stress Response And Immunity In Head And Neck Cancer, Yang Li, Pedram Yadollahi, Fonma N Essien, Vasanta Putluri, Chandra Shekar R Ambati, Karthik Reddy Kami Reddy, Abu Hena Mostafa Kamal, Nagireddy Putluri, Lama M Abdurrahman, Maria E Ruiz Echartea, Keenan J Ernste, Akshar J Trivedi, Jonathan Vazquez-Perez, William H Hudson, William K Decker, Rutulkumar Patel, Abdullah A Osman, Farrah Kheradmand, Stephen Y Lai, Jeffrey N Myers, Heath D Skinner, Cristian Coarfa, Kwangwon Lee, Antrix Jain, Anna Malovannaya, Mitchell J Frederick, Vlad C Sandulache

Faculty, Staff and Student Publications

Background: Exposomes are critical drivers of carcinogenesis. However, how they modulate tumor behavior remains unclear. Extensive clinical data show cigarette smoke to be a key exposome that promotes aggressive tumors, higher rates of metastasis, reduced response to chemoradiotherapy, and suppressed anti-tumor immunity. We sought to determine whether smoke itself can modulate aggressive tumor behavior in head and neck squamous cell carcinoma (HNSCC) through reprogramming of the cellular reductive state.

Methods: Using established human and murine HNSCC cell lines and syngeneic mouse models, we utilized conventional western blotting, steady state and flux metabolomics, RNA sequencing, quantitative proteomics and flow cytometry to …

A Phase Ii Trial Of Sitravatinib + Nivolumab After Progression On Immune Checkpoint Inhibitor In Patients With Metastatic Clear Cell Rcc, Andrew W Hahn, Nabil Adra, Ulka Vaishampayan, Lianchun Xiao, Nazli Dizman, Ying Yuan, Sagar S Mukhida, Matthew T Campbell, Jianjun Gao, Amado J Zurita, Eric Jonasch, Nizar M Tannir, Amishi Y Shah, Pavlos Msaouel

Faculty, Staff and Student Publications

Background: Sitravatinib, an oral multi-kinase inhibitor targeting VEGFR, TAM, and MET, has been shown to resensitize the tumor microenvironment to immune checkpoint inhibitors (ICI) by reducing immune-suppressive myeloid cells in metastatic clear cell RCC (ccRCC). ICI is the standard first-line (1L) treatment of metastatic ccRCC, and there is unmet need for improved treatment outcomes after progression on ICI. We hypothesized that sitravatinib plus nivolumab would revert an immunosuppressive tumor microenvironment (TME) to improve clinical outcomes.

Methods: In this investigator-initiated, phase II, multicenter trial (NCT04904302), patients with progressive metastatic ccRCC after 1-2 lines of treatment were enrolled into 3 …

Hif Regulates Multiple Translated Endogenous Retroviruses: Implications For Cancer Immunotherapy, Qinqin Jiang, David A Braun, Karl R Clauser, Vijyendra Ramesh, Nitin H Shirole, Joseph E Duke-Cohan, Nancy Nabilsi, Nicholas J Kramer, Cleo Forman, Isabelle E Lippincott, Susan Klaeger, Kshiti M Phulphagar, Vipheaviny Chea, Nawoo Kim, Allison P Vanasse, Eddy Saad, Teagan Parsons, Melissa Carr-Reynolds, Isabel Carulli, Katarina Pinjusic, Yijia Jiang, Rong Li, Sudeepa Syamala, Suzanna Rachimi, Eva K Verzani, Jonathan D Stevens, William J Lane, Sabrina Y Camp, Kevin Meli, Melissa B Pappalardi, Zachary T Herbert, Xintao Qiu, Paloma Cejas, Henry W Long, Sachet A Shukla, Eliezer M Van Allen, Toni K Choueiri, L Stirling Churchman, Jennifer G Abelin, Cagan Gurer, Gavin Macbeath, Richard W Childs, Steven A Carr, Derin B Keskin, Catherine J Wu, William G Kaelin

Faculty, Staff and Student Publications

Clear cell renal cell carcinoma (ccRCC), despite having a low mutational burden, is considered immunogenic because it occasionally undergoes spontaneous regressions and often responds to immunotherapies. The signature lesion in ccRCC is inactivation of the VHL tumor suppressor gene and consequent upregulation of the HIF transcription factor. An earlier case report described a ccRCC patient who was cured by an allogeneic stem cell transplant and later found to have donor-derived T cells that recognized a ccRCC-specific peptide encoded by a HIF-responsive endogenous retrovirus (ERV), ERVE-4. We report that ERVE-4 is one of many ERVs that are induced by HIF, translated …

Methylome-Wide Association Analyses Of Lipids And Modifying Effects Of Behavioral Factors In Diverse Race And Ethnicity Participants, Yao Hu, Jeff Haessler, Jessica I Lundin, Burcu F Darst, Eric A Whitsel, Megan Grove, Weihua Guan, Rui Xia, Mindy Szeto, Laura M Raffield, Scott Ratliff, Yuxuan Wang, Xuzhi Wang, Alison E Fohner, Megan T Lynch, Yesha M Patel, S Lani Park, Huichun Xu, Braxton D Mitchell, Joshua C Bis, Nona Sotoodehnia, Jennifer A Brody, Bruce M Psaty, Gina M Peloso, Michael Y Tsai, Stephen S Rich, Jerome I Rotter, Jennifer A Smith, Sharon L R Kardia, Alex P Reiner, Leslie Lange, Myriam Fornage, James S Pankow, Mariaelisa Graff, Kari E North, Charles Kooperberg, Ulrike Peters

Faculty, Staff and Student Publications

Circulating lipid concentrations are clinically associated with cardiometabolic diseases. The phenotypic variance explained by identified genetic variants remains limited, highlighting the importance of searching for additional factors beyond genetic sequence variants. DNA methylation has been linked to lipid concentrations in previous studies, although most of the studies harbored moderate sample sizes and exhibited underrepresentation of non-European ancestry populations. In addition, knowledge of nongenetic factors on lipid profiles is extremely limited. In the Population Architecture Using Genomics and Epidemiology (PAGE) Study, we performed methylome-wide association analysis on 9,561 participants from diverse race and ethnicity backgrounds for HDL-c, LDL-c, TC, and TG …

Chronic Viral Mimicry Induction Following P53 Loss Promotes Immune Evasion, Charles A Ishak, Sajid A Marhon, Naïri Tchrakian, Anjelica Hodgson, Helen Loo Yau, Isabela M Gonzaga, Melanie Peralta, Ilinca M Lungu, Stephanie Gomez, Sheng-Ben Liang, Shu Yi Shen, Raymond Chen, Jocelyn Chen, Biji Chatterjee, Kevin N Wanniarachchi, Junwoo Lee, Nicholas Zehrbach, Amir Hosseini, Parinaz Mehdipour, Siyu Sun, Alexander Solovyov, Ilias Ettayebi, Kyle E Francis, Aobo He, Taiyi Wu, Shengrui Feng, Tiago Da Silva Medina, Felipe Campos De Almeida, Jane Bayani, Jason Li, Spencer Macdonald, Yadong Wang, Sarah S Garcia, Elisa Arthofer, Noor Diab, Aneil Srivastava, Paul Tran Austin, Peter J B Sabatini, Benjamin D Greenbaum, Catherine A O'Brien, Trevor G Shepherd, Ming Sound Tsao, Katherine B Chiappinelli, Amit M Oza, Blaise A Clarke, Robert Rottapel, Stephanie Lheureux, Daniel D De Carvalho

Faculty, Staff and Student Publications

Epigenetic therapies facilitate transcription of immunogenic repetitive elements that cull cancer cells through ‘viral mimicry’ responses. Paradoxically, cancer-initiating events also facilitate transcription of repetitive elements. Contributions of repetitive element transcription towards cancer initiation, and the mechanisms by which cancer cells evade lethal viral mimicry responses during tumor initiation remain poorly understood. In this report, we characterize premalignant lesions of the fallopian tube along with syngeneic epithelial ovarian cancer models to explore the earliest events of tumorigenesis following loss of the p53 tumor suppressor protein. We report that p53 loss permits transcription of immunogenic repetitive elements and chronic viral mimicry activation …

Conformal Predictive Intervals In Survival Analysis: A Resampling Approach, Jing Qin, Jin Piao, Jing Ning, Yu Shen

Faculty, Staff and Student Publications

The distribution-free method of conformal prediction has gained considerable attention in computer science, machine learning, and statistics. Candès et al. extended this method to right-censored survival data, addressing right-censoring complexity by creating a covariate shift setting, extracting a subcohort of subjects with censoring times exceeding a fixed threshold. Their approach only estimates the lower prediction bound for type I censoring, where all subjects have available censoring times regardless of their failure status. In medical applications, we often encounter more general right-censored data, observing only the minimum of failure time and censoring time. Subjects with observed failure times have unavailable censoring …

Computational Sensitivity Evaluation Of Ultrasound Neuromodulation Resolution To Brain Tissue Sound Speed With Robust Beamforming, Boqiang Fan, Wayne Goodman, Sameer A Sheth, Richard R Bouchard, Behnaam Aazhang

Faculty, Staff and Student Publications

Low-intensity focused ultrasound (LIFU) neuromodulation requires precise targeting and high resolution enabled by phased array transducers and beamforming. However, focusing optimization usually relies on phantom measurements or simulations with inaccurate acoustic properties to degrade neuromodulation resolution. Therefore, this work analyzes the sensitivity of neuromodulation resolution, measured by off-target activation area (OTAA), to brain tissue sound speed. A Robust Optimal Resolution (ROR) beamforming method is proposed to minimize the worst-case OTAA with restricted sound speed inaccuracy and propagation information estimated with deviated sound speed. The propagation estimation model utilizes equivalent source method (ESM) to map sound field between different acoustic parameter …

Identifiability And Model Selection Frameworks For Models Of High-Grade Glioma Response To Chemoradiation, Khushi C Hiremath, Kenan Atakishi, Ernesto A B F Lima, Maguy Farhat, Bikash Panthi, Holly Langshaw, Mihir D Shanker, Wasif Talpur, Sara Thrower, Jodi Goldman, Caroline Chung, Thomas E Yankeelov, David A Hormuth Ii

Faculty, Staff and Student Publications

We have developed a family of biology-based mathematical models of high-grade glioma (HGG), capturing the key features of tumour growth and response to chemoradiation. We now seek to quantify the accuracy of parameter estimation and determine, when given a virtual patient cohort, which model was used to generate the tumours. In this way, we systematically test both the parameter and model identifiability. Virtual patients are generated from unique growth parameters whose growth dynamics are determined by the model family. We then assessed the ability to recover model parameters and select the model used to generate the tumour. We then evaluated …

Ven In Combination With 10-Day Dec In Newly Diagnosed Elderly Or Relapsed/Refractory Acute Myeloid Leukemia, And High-Risk Myelodysplastic Syndrome: Long Term Follow-Up Of A Phase 2 Trial, Mahesh Swaminathan, Courtney D Dinardo, Abhishek Maiti, Naveen Pemmaraju, Maro Ohanian, Navel G Daver, Guillermo Garcia-Manero, Ghayas C Issa, Gautam Borthakur, Farhad Ravandi, Guillermo Montalban-Bravo, Tapan M Kadia, Yesid Alvarado, Elias J Jabbour, Nicholas J Short, William G Wierda, Nitin Jain, Steven M Kornblau, Lucia Masarova, Sherry A Pierce, Wei Qiao, Jing Ning, Hagop Kantarjian, Marina Y Konopleva

Faculty, Staff and Student Publications

No abstract provided.

Long-Term Patient-Reported Bowel And Urinary Quality Of Life In Patients Treated With Intensity-Modulated Radiotherapy Versus Intensity-Modulated Proton Therapy For Localized Prostate Cancer, Kimberly R Gergelis, Miao Bai, Jiasen Ma, David M Routman, Bradley J Stish, Brian J Davis, Thomas M Pisansky, Thomas J Whitaker, Richard Choo

Faculty, Staff and Student Publications

Purpose: This study aimed to compare long-term patient-reported outcomes in bowel and urinary domains between intensity-modulated radiotherapy (IMRT) and intensity-modulated proton therapy (IMPT) for localized prostate cancer.

Methods and materials: Patients with clinical T1-T2 prostate cancer receiving IMRT or IMPT at a tertiary cancer center from 2015-2018 were analyzed to determine the changes in the prospectively collected bowel function (BF), urinary irritative/obstructive symptoms (UO), and urinary incontinence (UI) domains of EPIC-26. The mean changes in EPIC-26 scores were evaluated from pretreatment to 24 months post-radiotherapy for each modality. A score change >50% of the baseline standard deviation was considered a …

Prdm1 Is A Key Regulator Of The Nkt-Cell Central Memory Program And Effector Function, Gengwen Tian, Gabriel A Barragan, Hangjin Yu, Claudia Martinez-Amador, Akshaya Adaikkalavan, Xavier Rios, Linjie Guo, Janice M Drabek, Osmay Pardias, Xin Xu, Antonino Montalbano, Chunchao Zhang, Yanchuan Li, Amy N Courtney, Erica J Di Pierro, Leonid S Metelitsa

Faculty, Staff and Students Publications

Natural killer T cells (NKTs) are a promising platform for cancer immunotherapy, but few genes involved in the regulation of NKT therapeutic activity have been identified. To find regulators of NKT functional fitness, we developed a CRISPR/Cas9-based mutagenesis screen that uses a guide RNA (gRNA) library targeting 1,118 immune-related genes. Unmodified NKTs and NKTs expressing a GD2-specific chimeric antigen receptor (GD2.CAR) were transduced with the gRNA library and exposed to CD1d+ leukemia or CD1d-GD2+ neuroblastoma cells, respectively, over six challenge cycles in vitro. Quantification of gRNA abundance revealed enrichment of PRDM1-specific gRNAs in both NKTs and GD2.CAR NKTs, a result …

Improving The Oral Pathology Referral Process In An Academic Dental Institution, Gregory Olson

Translational Projects (Open Access)

Introduction: Inefficiencies in the oral cancer referral process pose serious risks, including delayed diagnoses and poor patient outcomes. This project aimed to improve the reliability and efficiency of referrals between dental students and oral pathology residents for patients with suspected oral cancer. Although early detection is widely acknowledged as critical, significant process breakdowns were identified, including the absence of a standardized workflow, inconsistent follow-up practices, and a lack of accountability and tracking mechanisms. As a result, approximately half of the patients recommended for biopsy did not undergo the procedure, leading to delayed diagnoses and potentially worse prognoses.

Methodology: …

Leveraging Observational And Rct Data For Understanding Interventions’ Efficacy: Applications In Progressive And Acute Neurological Diseases, Yaobin Ling

Dissertations and Theses (Open Access)

The advancement of drug repurposing for progressive and acute neurological diseases is hampered by the limitations of randomized clinical trials (RCTs) and observational data. This dissertation presents a comprehensive framework to integrate data-driven insights from multiple sources, including observational studies, RCTs, and synthetic data generation, to overcome these challenges.

The first study focuses on estimating treatment effects on the population level, which investigates the effects of routine and high-dose influenza vaccines on the risk of Alzheimer’s Disease and Related Dementias (ADRD) through a trial emulation framework applied to health claims data, addressing biases inherent in observational studies. The second study …

Standardizing Social Determinants Of Health Factors From Heterogeneous Sources To Improve Data Fairness, Yifang Dang

Dissertations and Theses (Open Access)

Social Determinants of Health (SDoH) significantly influence health outcomes, yet their representation in computational models remains fragmented. This dissertation addresses this gap by constructing an SDoH ontology (SDoHO), leveraging it to improve large language model (LLM) performance, and using LLMs to extract new knowledge and enrich the ontology. The overarching goal is to establish a feedback loop where ontology development enhances LLM-based extraction, and LLM-derived insights refine and expand the ontology. The methodology is structured across three aims: (1) ontology construction, (2) ontology-assisted LLM enhancement, and (3) LLM-driven ontology enrichment, with a focus on Alzheimer’s Disease and Related Dementias (ADRD). …

Computer-Aided Integrated Rehabilitation For Post-Stroke Patients Using Deep Learning Techniques, Kaichen Tang

Dissertations and Theses (Open Access)

Stroke is a leading cause of long-term disability, often requiring intensive rehabilitation and frequent clinical assessments such as the Fugl-Meyer Assessment. However, traditional in-person evaluations are resource-heavy and difficult to scale, limiting access for many patients. Additionally, monitoring vital signs like blood pressure and oxygen saturation typically depends on specialized equipment, leading to fragmented care and incomplete recovery insights. Even when assessments and monitoring are available, patients often struggle to maintain high-quality, intensive exercise routines without supervision, further limiting rehabilitation outcomes.

To address these challenges, this dissertation proposes an AI-driven, smartphone-based framework that integrates automated motor function assessment, non-invasive vital …

Leveraging And Advancing The Foundation Models For Clinical Trajectory Analysis Based On Electronic Health Record Data, Jianping He

Dissertations and Theses (Open Access)

Enhancements in foundation models have provided significant potentials for clinical applications. This dissertation leverages and advances foundation models for a range of clinical tasks using electronic health record (EHR) data, demonstrating their applicability in Clinical Temporal Relation Extraction (CTRE), disease prediction, and patients trajectory analysis.

Aim 1 is to efficiently adapt large language models (LLMs) for CTRE in both full data and few-shot settings. This study leveraged four LLMs: GatorTron-Base, GatorTron-Large, LLaMA3.1, and MeLLaMA. The proposed fine-tuning strategies include: (a) standard fine-tuning; (b) hard-prompting; (c) soft-prompting; (d) Low-Rank Adaptation (LoRA). We found that nearly all proposed fine-tuning strategies outperformed existing …

Precision Drug Dosing And Disease Risk Prediction: Advanced Deep Learning Artificial Intelligence Approaches For Precision Medicine Using Structured Electronic Health Records Data, Bingyu Mao

Dissertations and Theses (Open Access)

Advancements in precision medicine increasingly rely on data-driven approaches to improve clinical decision-making. This work presents three interconnected studies that leverage deep learning, reinforcement learning, and comparative modeling to address key challenges in precision drug dosing and disease risk prediction. Together, these contributions offer novel computational frameworks that enhance model performance, improve dosing strategies, and guide model selection for clinical predictive tasks.

To improve individualized vancomycin therapeutic drug monitoring (TDM), the first study introduces PKRNN-2CM, a novel deep learning framework that integrates a two-compartment pharmacokinetic (PK) model with recurrent neural networks (RNNs). While one-compartment models are commonly used in clinical …

Computational Methods For Enhancing The Quality And Interoperability Of Biomedical Terminologies, Xubing Hao

Dissertations and Theses (Open Access)

Biomedical ontologies or terminologies not only serve as a part of the metadata standards for describing data in the FAIR Data Principles (Findable, Accessible, Interoperable, Reusable), but also play a vital role in downstream applications such as cohort identification from electronic health records (EHR). However, there are two critical barriers that may lead to ambiguity, complexity, and inaccuracies in such ontology-based downstream applications. The first barrier is the quality of the ontology. Despite efforts by ontology curators to ensure ontology accuracy and comprehensiveness, errors and inconsistencies are unavoidable. The second barrier is the semantic heterogeneity since human experts may use …

Advancing Genetic Association Discovery In Brain Mri Using Unsupervised And Self-Supervised Deep Learning: Exploring Learning Dynamics, Region-Specific Features, And Spatially Resolved Representations, Sheikh Muhammad Saiful Islam

Dissertations and Theses (Open Access)

Deep learning has unlocked significant potential for advancing the discovery of genetic associations in imaging genetics, particularly in brain imaging using T1-weighted Magnetic Resonance Imaging (MRI). Traditional methods for this task often relied on hand-crafted or semi-automated feature extraction, followed by genetic association studies utilizing these features. While effective, these approaches are limited by their lack of data-driven exploration and the generation of features with limited informativeness.

Unsupervised deep learning, in particular, has emerged as a powerful alternative, addressing some of these limitations by enabling automated and data-driven feature discovery. Recent research in this domain has primarily focused on adapting …

Harnessing Knowledge And Data For Clinical Information Extraction In The Era Of Large Language Models, Yan Hu

Dissertations and Theses (Open Access)

The rapid digitization of healthcare records has led to the widespread adoption of Electronic Health Records (EHRs), which contain rich, unstructured clinical notes. These notes hold valuable information for patient care and clinical research, but their complexity and unstructured nature pose significant challenges for effective utilization. Clinical Information Extraction (IE) aims to bridge this gap by transforming unstructured text into structured data, enabling automated analysis. Traditional Natural Language Processing (NLP) techniques, such as Named Entity Recognition (NER), have been widely used for clinical IE, but recent advancements in Large Language Models (LLMs) like GPT-3.5, GPT-4, and LLaMA have opened new …

Accelerating Drug Repurposing And Target Discovery Using Deep Learning, Kohong Lin

Dissertations and Theses (Open Access)

Drug discovery is a long-lasting and expensive process. Computational approaches, particularly deep learning techniques, offer the potential to accelerate this process by integrating diverse perspectives from drug discovery theories and capturing intricate patterns within large, multimodal datasets. This dissertation explores deep learning methodologies to accelerate drug repurposing and genetic target discovery. The first aim focuses on integrating multimodal data, including chemical structures, disease genetics, and systems biology, into comprehensive disease knowledge graphs, followed by applying graph neural networks (GNNs) to prioritize repurposable drug candidates. The second aim is to develop a heterogeneous GNN-based approach capable of modelling distinct semantic relationships …

An Interoperability Framework Across Legacy Emr Systems, Sheena Rogers

Translational Projects (Open Access)

The 21st Century Cures Act (H.R.34 – 114th Congress, 2016) specifies that patients and authorized providers must be able to receive copies of their medical records with standard medical history details upon request. While the data archival system, DRIS (Data Retention Interoperability Solution), has comprehensive patient summary reports centralizing over 100 legacy electronic health records (EHR) with broad data elements, it is not 100% compliant with the upcoming 21st Century Cures Act’s interoperability standards. As a result, patients and providers are not receiving copies of their medical records with all of the required medical history and fields in this new …

An Examination Of Ambulatory Care Code Specificity Utilization In Icd-10-Cm Compared To Icd-9-Cm: Implications For Icd-11 Implementation, Susan H Fenton, Cassandra Ciminello, Vickie M Mays, Mary H Stanfill, Valerie Watzlaf

Faculty, Staff and Student Publications

Objective: The ICD-10-CM classification system contains more specificity than its predecessor ICD-9-CM. A stated reason for transitioning to ICD-10-CM was to increase the availability of detailed data. This study aims to determine whether the increased specificity contained in ICD-10-CM is utilized in the ambulatory care setting and inform an evidence-based approach to evaluate ICD-11 content for implementation planning in the United States.

Materials and methods: Diagnosis codes and text descriptions were extracted from a 25% random sample of the IQVIA Ambulatory EMR-US database for 2014 (ICD-9-CM, n = 14 327 155) and 2019 (ICD-10-CM, n = 13 062 900). Code …

Revisions To The Safety Assurance Factors For Electronic Health Record Resilience (Safer) Guides To Update National Recommendations For Safe Use Of Electronic Health Records, Dean F Sittig, Trisha Flanagan, Patricia Sengstack, Rosann T Cholankeril, Sara Ehsan, Amanda Heidemann, Daniel R Murphy, Hojjat Salmasian, Jason S Adelman, Hardeep Singh

Faculty, Staff and Student Publications

The Safety Assurance Factors for Electronic Health Record (EHR) Resilience (SAFER) Guides provide recommendations to healthcare organizations for conducting proactive self-assessments of the safety and effectiveness of their EHR implementation and use. Originally released in 2014, they were last updated in 2016. In 2022, the Centers for Medicare and Medicaid Services required their annual attestation by US hospitals.

Objectives: This case study describes how SAFER Guide recommendations were updated to align with current evidence and clinical practice.

Materials and methods: Over nine months, a multidisciplinary team updated SAFER Guides through literature reviews, iterative feedback, and online meetings.

Results: We reduced …

Llm-Ie: A Python Package For Biomedical Generative Information Extraction With Large Language Models, Enshuo Hsu, Kirk Roberts

Faculty, Staff and Student Publications

Objectives: Despite the recent adoption of large language models (LLMs) for biomedical information extraction (IE), challenges in prompt engineering and algorithms persist, with no dedicated software available. To address this, we developed LLM-IE: a Python package for building complete IE pipelines.

Materials and methods: The LLM-IE supports named entity recognition, entity attribute extraction, and relation extraction tasks. We benchmarked it on the i2b2 clinical datasets.

Results: The sentence-based prompting algorithm resulted in the best 8-shot performance of over 70% strict F1 for entity extraction and about 60% F1 for entity attribute extraction.

Discussion: We developed a Python package, LLM-IE, …