Exploring Agentic AI and Multi-Agent Systems for ICU Applications
Name:
Dr. Azra Bihorac
Email
abihorac@ufl.edu
Phone
(352) 273-5995
Faculty Department/Division
Nephrology, Hypertension & Renal Transplantation
This project is primarily:
Literature Review
Research Project Description:
This research and development project focuses on leveraging Agentic AI and multi-agent systems to enhance decision-making processes in Intensive Care Units (ICUs). The project aims to develop a foundational understanding of how autonomous agents can collaborate to optimize patient outcomes, improve resource allocation, and support healthcare providers in high-pressure environments.
The medical student will work closely with faculty mentors to investigate existing applications of multi-agent systems in healthcare, identify gaps in current methodologies, and propose innovative solutions. The hypothesis centers on the potential for agentic AI systems to provide timely and accurate support in critical care settings. Methods will include literature reviews, system design, and potential development of simulated agent interactions to evaluate performance. The student’s role will include contributing to the formulation of research questions, assisting with data collection and analysis, and participating in the design and evaluation of the multi-agent system.
Does this project have an international component or travel?
No
Optimizing Hypotension Management in Critical Care Using Deep Reinforcement Learning
Name:
Dr. Tezcan Ozrazgat-Baslanti
Email
Tezcan.OzrazgatBaslanti@medicine.ufl.edu
Phone
(352) 273-6668
Faculty Department/Division
Nephrology, Hypertension & Renal Transplantation
This project is primarily:
Clinical
Research Project Description:
Research Project Description: Patients and physicians make essential decisions on which diagnostic and therapeutic interventions should be performed or deferred under time constraints and uncertainty regarding patients’ diagnoses and predicted response to treatment which may lead to cognitive and judgment errors. Current practices in the administration of intravenous fluids and vasopressors during ICU stay as a treatment of hypotension are suboptimal. Reinforcement learning has significant potential to complement clinical decision-making by enabling sound judgment and optimal treatment strategies in critical care settings. We aim to develop a deep reinforcement learning model that delivers individualized and clinically interpretable treatment decisions to manage blood pressure, reduce hypotension associated complications, and improve outcomes for critically ill patients. The role of the medical student in this project will be to assist with literature review, manuscript preparation, and assist with project tasks.
Does this project have an international component or travel?
No
Integrating Artificial Intelligence into Prostate Cancer Diagnostics: A Hands-On Learning Experience for Medical Students
Name:
Dr. Wei Shao
Email
weishao@ufl.edu
Phone
(352) 273-5354
Faculty Department/Division
Nephrology, Hypertension & Renal Transplantation
This project is primarily:
Translational
Research Project Description:
Artificial Intelligence (AI) is transforming the landscape of medical diagnostics, offering potential improvements in speed and accuracy that could significantly enhance patient care. In prostate cancer screening, AI has shown great promise in improving the detection of clinically significant cancers, especially when integrated with advanced imaging modalities like micro-ultrasound. This project aims to provide medical students with a hands-on learning experience, allowing them to apply AI techniques to assist in prostate cancer diagnosis. The hypothesis is that by engaging medical students in developing and refining AI models, they will acquire valuable skills in both medical imaging and AI, enabling them to better understand and utilize emerging technologies in clinical practice.
Medical students participating in this project will work under the mentorship of Dr. Wei Shao, learning to preprocess micro-ultrasound imaging data, train convolutional neural networks (CNNs) and vision transformers, and evaluate AI model performance using metrics such as sensitivity, specificity, and ROC-AUC. Their responsibilities will also include testing the integration of non-imaging biomarkers, such as PSA levels and patient demographics, into multimodal AI systems. Through these experiences, students will learn how AI can enhance diagnostic workflows and address clinical challenges. This project will equip medical students with cutting-edge skills in AI and medical imaging, positioning them to contribute to the future of AI-driven healthcare.
Below are some relevant publications from Dr. Shao’s lab.
- Imran, M., Nguyen, B., Pensa, J., Falzarano, S.M., Sisk, A.E., Liang, M., DiBianco, J.M., Su, L.M., Zhou, Y., Brisbane, W.G., and Shao, W. Image Registration of In Vivo Micro-Ultrasound and Ex Vivo Pseudo-Whole Mount Histopathology Images of the Prostate: A Proof-of-Concept Study. Biomedical Signal Processing and Control, 2024.
- Shao, W., Vesal, S., Soerensen, S. J. C., Bhattacharya, I., Golestani, N., Yamashita, R., Kunder, C. A., Fan, R. E., Ghanouni, P., Brooks, J. D., Sonn, G. A., and Rusu, M. RAPHIA: A deep learning pipeline for the registration of MRI and whole-mount histopathology images of the prostate. Computers in Biology and Medicine, 2024.
- Jiang, H., Imran, M., Muralidharan, P., Patel, A., Pensa, J., Liang, M., Benidir, T., Grajo, J.R., Joseph, J.P., Terry, R., DiBianco, J.M., Su, L.M., Zhou, Y., Brisbane, W.G., and Shao, W. MicroSegNet: A Deep Learning Approach for Prostate Segmentation on Micro-Ultrasound Images. Computerized Medical Imaging and Graphics, 2024.
- Bhattacharya, I., Seetharaman, A., Kunder, C., Shao, W., Chen, L. C., Soerensen, S. J. C., Wang, J. B., Teslovich, N. C., Fan, R. E., Ghanouni, P., Brooks, J. D., Sonn, G. A., and Rusu, M. Selective identification and localization of indolent and aggressive prostate cancers via CorrSigNIA: an MRI-pathology correlation and deep learning framework. Medical Image Analysis, 2022.
- Shao, W., Banh, L., Kunder, C.A., Fan, R.E., Soerensen, S.J.C., Wang, J.B., Teslovich, N.C., Madhuripan, N., Jawahar, A., Ghanouni, P., Brooks, J.D., Sonn, G.A., and Rusu, M. ProsRegNet: A deep learning framework for registration of MRI and histopathology images of the prostate. Medical Image Analysis, 2021.
Does this project have an international component or travel?
No