The Mississippi IDeA Network of Biomedical Research Excellence (INBRE) cordially invites faculty members at Mississippi’s Primarily Undergraduate Institutions (PUIs) to attend the 3rd Annual Mississippi INBRE Data Science Workshop, to be hosted on Thursday, March 19, 2026, in room D-2 at the Mississippi Coast Convention Center during the 90th Annual Mississippi Academy of Sciences (MAS) Meeting. The workshop will begin at noon and conclude at 1 PM.

Focused on demonstrating how data science and AI can be used to enhance research, workshop speakers will introduce data science and AI applications and innovations across various domains and highlight a beginner-friendly machine learning tool that everyone can access.

This workshop is free and open to all PUI faculty attendees of the 90th Annual MAS Meeting. Pre-register by Friday, February 27, to receive a complimentary lunch voucher to redeem at the meeting facility concession stand!

To be eligible to attend the workshop, you must be a faculty member at a PUI in Mississippi and a registered attendee of the 90th Annual MAS Meeting. Faculty who pre-register by the February 27 deadline will receive a complimentary lunch voucher and may also pre-register up to two student guests, each of whom will also receive a complimentary lunch voucher. Pre-registration is not required to attend the workshop, however, faculty must pre-register by the deadline to receive a complimentary lunch voucher. All attendees are welcome to bring lunch while enjoying the presentations and discussions.

Pre-registration closes Friday, February 27, 2026, or when session fills.

Click the link below to register now!

Register

Presentations

“Novel Medical AI: Preterm Labor Prediction”

Invited Speaker: Benjamin F. Dribus, Ph.D.

Director of Artificial Intelligence, Chair of Mathematics, Medical Research Associate, William Carey University

Abstract: Preterm labor remains difficult to predict reliably and plays a significant role in maternal and neonatal health outcomes. Moderate to severe preterm labor (PTL), defined in the present study as labor onset before 34 weeks gestational age (GA), is associated with high incidences of adverse events. Roughly 5% of non-induced pregnancies examined in the present study ended in PTL, signifying hundreds of thousands of cases nationally per year. While interventions to address PTL can be effective, many cases remain unanticipated and fail to receive timely treatment.  This suggests that PTL often depends on complex multivariate interactions that elude simple algorithms.  Artificial intelligence (AI) offers promising methods to address this problem.  In this study, we introduce an individualized, noninvasive, cost-effective AI-based PTL prediction tool to enhance current standards of care. This tool leverages ensembles of novel neural networks to output a running risk index optimized to each individual patient’s symptoms, risk factors, and stage of pregnancy. It consistently outperforms simpler statistical and machine learning methods under equivalent conditions, achieving roughly 83% balanced accuracy on the validation data.

Bio: Dr. Dribus is Director of Artificial Intelligence at William Carey University, where he also serves as Chair of the Math Department and Research Associate with the College of Osteopathic Medicine. He holds a patent on a novel AI architecture called sparse local highly connected artificial neural networks (SLC networks). He is a member of the Foundational Questions Institute (FQxI). He recently obtained an NIH funded grant through the MS-INBRE network to develop medical applications of his invention.

“From Data to Discovery: Computational Mining of Chemical Datasets for Consumption Safety and Forensic Pharmacology”

 Invited Speaker: Scoty Hearst, Ph.D.

Assistant Professor, Department of Chemistry and Biochemistry, Mississippi College

Abstract: AI is transforming modern STEM by enhancing research, education, and automating tasks. AI can aid in scientific discovery through data analysis. This allows for greater innovation in scientific fields. We are currently using AI and computational analysis to find patterns and trends in large analytical chemistry data sets. Using WEKA software coupled with toxic metal analysis, we were able to answer questions concerning consumption safety of fish from the Mississippi River and gluten-free cassava root products. These data sets analyzed by AI revealed how toxic metals influence consumption safety. GCMS analysis is a common tool used in forensic pharmacology to identify drugs, poisons, and their metabolites in biological samples. However, GCMS analysis often results in thousands of compounds per sample making analysis an enormous task. Using ChatGPT coupled with GCMS analysis of forensic samples, we were able to identify medical related compounds in human remains to determine medical treatment strategies and cause of death. Overall, computational analysis and AI can streamline data-intensive tasks to enhance discovery of patterns and trends in large analytical chemistry data sets.

Bio: Dr. Scoty Hearst received his PhD in Biochemistry from University of Mississippi Medical Center. He is currently an Assistant Professor at Mississippi College in the Chemistry and Biochemistry Department.

Dr. Hearst’s research uses analytical chemistry and biochemistry techniques to connect wildlife and the environment to public health. His projects including zoonotic diseases surveillance in Mississippi wildlife and surveillance of environmental contaminates of human health concern in aquatic and terrestrial species and their environments.

“A Quick Start to Machine Learning”

Invited Speaker: Jingyi “Catherine” Shi, Ph.D.

Data Science Core Associate Director, Mississippi INBRE

Assistant Professor of Statistics, Department of Mathematics and Statistics, Mississippi State University

Abstract: WEKA is an open-source, GUI-based machine learning tool developed by the University of Waikato, New Zealand. Dr. Shi previously introduced WEKA during the 1^st Data Science Workshop at the 88^th Annual MAS Meeting and received highly positive feedback. Due to continued interest in the tool, she will reintroduce WEKA at the 90^th Annual MAS Meeting for new attendees and anyone seeking a practical, beginner-friendly entry point to machine learning. WEKA is available for download at https://ml.cms.waikato.ac.nz/weka/.

Bio: Dr. Shi joined Mississippi State University as an Assistant Professor of Statistics in 2019, right after her doctoral study in Health Informatics at UNC Charlotte. Her research work spans a wide range of sub-areas in the Data Science field, including applied statistics, supervised machine learning, feature selection, knowledge extraction and representation, semantic web and ontology, applied natural language processing, quantification of data quality, and AI software development. She has published 13 peer-reviewed journal papers and six conference papers as of 2025, of which six journals are ranked Q1 by CiteScore. She also published an R package on CRAN called “CASMI.” As an educator, she has taught or developed courses in statistical methods, machine learning, and databases.

Workshop Organizers

Dr. Yufeng Zheng

Workshop Co-Chair

Data Science Core Director, Mississippi INBRE

Associate Professor, Department of Data Science, University of Mississippi Medical Center

Bio: Dr. Yufeng Zheng is an Associate Professor of Data Science at the University of Mississippi Medical Center and Director of the Data Science Core for the NIH-funded MS INBRE initiative. He teaches graduate courses in Advanced Machine Learning and Deep Learning Applications and has received UMMC’s 2022 Teaching Excellence Award.  He is the principal investigator of many funded projects and received 2025 UMMC’s Research Excellence Award and Discovery Award. He holds a patent in facial recognition, has published three books and more than 90 peer-reviewed papers, and is an NVIDIA-certified Deep Learning Instructor, a Cisco Certified Network Professional (CCNP), and a senior member of IEEE and SPIE. His research centers on image analysis, machine learning, AI, and computer-aided diagnosis.

Dr. Jingyi “Catherine” Shi

Workshop Co-Chair

Data Science Core Associate Director, Mississippi INBRE

Assistant Professor of Statistics, Department of Mathematics and Statistics, Mississippi State University

Bio: Dr. Shi joined Mississippi State University as an Assistant Professor of Statistics in 2019, right after her doctoral study in Health Informatics at UNC Charlotte. Her research work spans a wide range of sub-areas in the Data Science field, including applied statistics, supervised machine learning, feature selection, knowledge extraction and representation, semantic web and ontology, applied natural language processing, quantification of data quality, and AI software development. She has published 13 peer-reviewed journal papers and six conference papers as of 2025, of which six journals are ranked Q1 by CiteScore. She also published an R package on CRAN called “CASMI.” As an educator, she has taught or developed courses in statistical methods, machine learning, and databases.