The goal of this research group is to apply computational methods and technologies from the Computer and Data Science domains to various applications in healthcare and Medicine to support clinical and public health decisions to improve the health of individuals and populations, narrow health disparities, decrease healthcare costs, and improve overall human well-being.
We collaborate with several organizations, including the Population Health Intelligence group within the Oak Ridge National Lab’s Center for Biomedical Informatics at the University of Tennessee Health Science Center, through which we design research pipelines for integrating proximal/ downstream clinical with distal/upstream non-clinical risk factors/data and building machine learning prediction models to study several health outcomes. We also incorporate results from those models into intelligent digital health data management and surveillance platforms.
We also work with the Ochsner Xavier Institute for Health Equity and Research (OXIHER), and outcomes research group at Ochsner health where we provide informatics and analytics expertise for advanced analytics for health system-level, data-driven, clinical initiatives.
eXplainable AI
We leverage knowledge graphs as explainable models.
Fight for the things that you care about, but do it in a way that will lead others to join you..
Ruth Bader Ginsburg
Precision Population Health Early Detection: building Incorporating ML algorithms into building predictive prognostic models that study healthcare data retrospectively
Precision Population Health Intervention: Health Education and Promotion using recommender and question answering systems powered by personal digital health libraries. Currently ongoing projects:
Digital Health Monitoring Tools: Results from ML models can inform future decisions by incorporating them as metrics into existing tools or proof of concept prototypes. Two projects that resulted in publications:
Knowledge Representation and reasoning.
Graph representation learning
Applied Machine Learning
Privacy preserving Machine Learning
Health Education and Promotion
Applications of eXplainable AI.
Current Projects
| Precision Population Health Early Detection Project 1.a: ML Powered Knowledge Discovery for Precision Maternal Fetal Medicine: this project is in collaboration with Dr. Price-Haywood Dr. Olet, Dr. Howard, OXIHER , Ochsner Health.  Project 1.b: Privacy-preserving ML analysis using Homomorphic Encryption: this project is in collaboration with cyber security research group lead by Dr. Baksi, at Illinois State University. |
| Project 2: Predicting outcomes in Pediatric Asthma based on risk factors identified in literature [5]. This project is a collaboration with Dr. Espinoza, Children’s Hospital of Los Angeles |
| Precision Population Health : |
Project 3: A Digital Personal Health Coaching Platform for HPV Vaccine Promotion and Education [12]. This project is a collaboration with Dr. Shaban Nejad, University of Tennessee Health Science Center. |
| AI-driven Clinical Decision Support (CDS) |
| Project 4: Cloud-based platform for ICU patient monitoring. This project is collaboration with Dr. Rezgui, Illinois State University. Project 4.a: Using in hospital data: low and high quantile ranges for predicting Vital Signs in ICU patients [7,8] Project 4.b: Using out of hospital data for ICU patient monitoring. |
| Project 5: Enhanced emergency crowding score within Epic EHR system [1] Project 5.a: OEDOCS v2.0 Project 5.b: Brining OEDOCS into realization |
| Project 6: X-CART (an eXplainable CAncer Radiation Therapy Platform): Knowledge graph based Deep Learning for predicting Access to care domain based risk factors. This project is collaboration with Dr. Shaban Nejad, University of Tennessee Health Science Center. |