Date: November 21, 2025
Joel Harley, PhD
Associate Professor, Kent and Linda Fuchs Faculty Fellow, University of Florida; Director, SmartData Lab
Aron Culotta, PhD
Associate Professor, School of Science & Engineering, Director, Center for Community-Engaged Artificial Intelligence, Tulane University
Date: February 20, 2026
Ravi Teja Bhupatiraju, MBBS, PhD, Health Informatics Research Scientist, Center for Applied Artificial Intelligence, University of Louisiana at Lafayette ravi-teja.bhupatiraju@louisiana.edu.
Topic: Data Science Approaches to Primary Care/Precision Medicine
Date: Friday, March 20, 2026
Guest Speaker: Lizheng Shi, PhD, MsPharm, M, School of Public Health and Tropical Medicine, Tulane University
Biography: Dr. Shi is the founding director of Tulane’s Health Systems Analytics Research Center (HSARC). He leads the research team that receives awards from professional organizations, including the 2025 Hartzema Distinguished Speaker Award, Best Research Paper Award from the Patient Access Network Foundation, and American Journal of Managed Care. He advised students for three of the best research presentation awards in the International Society for Pharmacoeonomics and Outcomes Research (2013, 2015, and 2024). He has published more than 300 papers in peer-reviewed journals and served as principal investigator and co-PI for more than 40 research grants and contracts from AHRQ, CDC, NIH, PCORI, and other public and private funding sources. Dr. Shi is dedicated to disseminating and translating population health knowledge at the local, national, and international levels. He is the associate editor of Value In Health and co-editor-in-chief for Pharmacoeconomics and Policy.
Dr. Shi’s current health services research interest focuses on innovative health technologies to improve healthcare quality, access, and cost of patient-centered care from the equity perspective, using pharmaco-economics, health technology assessment, health analytics, and policy evaluation. Dr. Shi has used data science tools (artificial intelligence and machine learning) to improve policy evaluation and analytics. He has conducted several projects on goal optimization and data analytics to support diabetes management including the Building, Relating, Assessing, and Validating Outcomes (BRAVO) diabetes simulation. The BRAVO model is the first American-based patient-level microsimulation model. The improvements in the BRAVO model included the US-based model fitting for the US general population of diabetes, global calibration to other countries, and the adaptation module for its application in the electronic medical records system. He has worked on several projects using big data analytics to further optimize treatment for diseases with multiple goals (e.g., HbA1c, lipid, and blood pressure). He has fostered extensive and inclusive partnerships with state government agencies, community nonprofit organizations, patients, health systems, federally qualified health centers, and payers.
Learn more
Topic: Ethical Sensemaking in AI Mental Health Chatbots
Date: Friday, April 17, 2026
Guest Speaker: Beenish Chaudhry, PhD, Associate Professor, School of Computing and Informatics, University of Louisiana at Lafayette
Biography: Dr. Chaudhry’s work focuses on human-computer interaction (HCI) and artificial intelligence (AI), exploring the intersection of technology, design, and user experience (UX). She is passionate about designing ethical, user-centered AI tools and improving the integration of AI into creative workflows and healthcare. She currently teaches courses on system design and analysis, HCI, and UX principles. She prefers hands-on techniques in her teaching, fostering an interactive and practical learning environment for her students. Her work continues to push the boundaries of AI in design and healthcare, ensuring that technology is accessible and beneficial to all users.
Abstract
As AI adoption expands, mental health chatbots are increasingly used for emotional support and self-management, yet their roles and responsibilities remain ethically ambiguous in everyday use. Prior research often treats these boundaries as externally defined, overlooking how users interpret and negotiate them in practice.
We analyze a large corpus of app store reviews of four widely used mental health chatbots using topic modeling and qualitative analysis. We identify three recurring processes: interactional role inference, ethical negotiation during success and breakdown, and collective boundary contestation through reviews.
We show that users infer roles from conversational cues and regulate trust as a bounded form of delegation. Breakdowns act as ethical inflection points, while reviews externalize judgments into collective boundary work and vernacular governance.
These findings demonstrate that ethical boundaries are actively constructed through interaction and platform participation, shifting design toward supporting ethical sensemaking as an ongoing, socially distributed process.
Topic: AI-Powered imaging technology for cancer detection
Date: Friday, August 21, 2026
Guest Speaker: Brian Summa, Associate Professor, School of Science & Engineering, Tulane University
Project Overview:
Brian Summa is co-leading a groundbreaking MAGIC-SCAN project, an AI-powered imaging technology designed to help surgeons detect even the smallest traces of cancer during surgery, reducing the need for repeat procedures and improving patient care.
The first, MAGIC-SCAN (Machine-learning Assisted Gigantic Image Cancer margin SCANner) combines extreme-field-of-view optical sectioning and super-resolution structured illumination microscopy (OS-SR-SIM) to obtain very high-speed virtual pathology imaging of cancer tumor margin surfaces at 2x the resolution allowed by diffraction.
Currently, it can take days to weeks before a surgeon knows whether all the tumor has been removed, their our goal is to get that down to 10 minutes, while the patient is still on the table.
Their work is helping ensure that world-class cancer innovation reaches patients everywhere, including rural communities where access to advanced care can be limited.
Research Interests:
His research interests are in large data processing, analysis, and interaction for data visualization and computer graphics applications. His work has supported the creation of deployable, large data processing software that scales from mobile devices to supercomputers. His research in visualization is highly interdisciplinary and has allowed him to work closely with researchers in a variety of disciplines such as geology, physics, and neurobiology.
Agata Rozga, PhD.
Principal Research Scientist in the School of Interactive Computing at the Georgia Institute of Technology, and the Director of Translational Research for AI-CARING
Topic: The Last Mile Begins in the First: Insights from Developing Technologies For Autism and Aging
Date: Friday, October 16, 2026
Time: 2:30 pm – 3:30 pm CT / 3:30 pm – 4:40 pm ET
Biography: Dr. Rozga’s research sits at the intersection of health, AI, and human-centered computing. She develops novel tools and methods that can he used to study health behaviors in real-world settings, with a particular emphasis on behaviorally defined, progressive conditions such as autism and cognitive decline. She aims to bridge the gap between translational research and scalable technologies that improve the lives of individuals, their care partners, and providers. Her work in support of early detection of autism was commercialized in collaboration with Behavior Imaging Solutions.
Dr. Rozga previously served as Head of Human Centered Design and Head of Product for Diligent Robotics, a startup developing hospital service robots to support nursing, which was recently acquired by Serve Robotics.
AI-CARING is an NSF AI Institute focused on developing AI-based interactive technologies to support aging in place.
Project Overview: In this talk, I propose that the successful translation of predictive models and interactive technologies from research into healthcare settings begins with decisions and assumptions made early in the R&D process: decisions in how foundational models are trained and what is being predicted, assumptions about end users and their needs, as well as the settings and metrics used to evaluate early prototypes. I draw upon my work in autism and aging, which spans foundational AI, real-world deployments, and commercialization, to illustrate key lessons about theoretical and practical considerations in developing and evaluating AI technologies that can successfully bridge the translational gap.