Four-Year ORBIL-AI Project Builds on Faculty Research Presented at Yale University to Explore How Artificial Intelligence Can Better Interpret Complex Biomedical Data
ORBIL-AI Project Overview and Funding Morgan State University has received a four-year, approximately $520,000 research award from the National Institutes of Health (NIH) through the National Library of Medicine (NLM). The project, titled ORBIL-AI: Orthogonality-Regularized Biomedical Information-theoretic Learning, commenced on September 10, 2026, and is set to conclude by July 31, 2030. Led by Dr. Timothy Oladunni, assistant professor in Morgan’s Department of Computer Science, the research aims to address the challenge of how artificial intelligence processes, integrates, and interprets complex biomedical information, specifically focusing on distinguishing complementary data from redundant information across multiple sources. The initial award was $130,134, with a total anticipated funding of $520,000 over the project's duration through NIH’s Support for Research Excellence (SuRE) R16 program. When More Data Does Not Necessarily Mean Better Results The scientific basis for ORBIL-AI stems from Dr. Oladunni's earlier work on redundancy and complementarity in multimodal machine learning, which he presented at Yale University. His research, including the Complementary Feature Domains (CFD) theory, highlighted that integrating additional feature domains into AI models can introduce overlapping information and increase computational demands without necessarily improving predictive performance. ORBIL-AI expands on this by investigating orthogonality-regularized, information-theoretic learning to develop mathematical and computational methods. This approach enables AI systems to more effectively learn complementary information by studying how computational models represent shared and distinct data across various biomedical observations, potentially improving model generalizability, interpretability, and calibration. Expanding Biomedical AI Research and Student Opportunities at Morgan This NIH award significantly strengthens Morgan State University's dedication to fostering investigator-led research, enhancing interdisciplinary collaboration, and creating valuable opportunities for students in emerging scientific fields. The ORBIL-AI project integrates information theory, mathematical representation learning, artificial intelligence, and biomedical signal analysis, thereby contributing to a growing body of research on advanced computational methods for complex scientific challenges. The initiative also builds upon Dr. Oladunni’s existing research in explainable machine learning, multimodal biomedical computing, and theoretical computational representations. Furthermore, it aligns with the NIH SuRE program's objectives to reinforce research capacity and promote meaningful student engagement in biomedical discovery, leveraging Dr. Oladunni's prior visiting faculty research fellowship at Yale University in cardiovascular science and computational AI in medicine. Advancing the Future of Artificial Intelligence Although ORBIL-AI focuses on biomedical signals, its core scientific questions about effectively integrating information from multiple sources have broad implications beyond healthcare. As AI systems increasingly depend on diverse data, a crucial challenge is to determine how each source contributes to a computational model’s understanding and whether additional information genuinely enhances its performance. By exploring the mathematical relationships between shared, redundant, and complementary information, ORBIL-AI aims to advance the theoretical foundations of multimodal machine learning. This research holds the potential to inform the development of more efficient, reliable, and interpretable AI systems, driving future innovations in biomedical research and other data-intensive scientific domains. For Morgan State University, this NIH-supported initiative represents a key step in expanding its contributions to foundational artificial intelligence research and preparing the next generation of scientists to tackle evolving technological challenges.
ORBIL-AI Project Overview and Funding
Morgan State University has received a four-year, approximately $520,000 research award from the National Institutes of Health (NIH) through the National Library of Medicine (NLM). The project, titled ORBIL-AI: Orthogonality-Regularized Biomedical Information-theoretic Learning, commenced on September 10, 2026, and is set to conclude by July 31, 2030. Led by Dr. Timothy Oladunni, assistant professor in Morgan’s Department of Computer Science, the research aims to address the challenge of how artificial intelligence processes, integrates, and interprets complex biomedical information, specifically focusing on distinguishing complementary data from redundant information across multiple sources. The initial award was $130,134, with a total anticipated funding of $520,000 over the project's duration through NIH’s Support for Research Excellence (SuRE) R16 program.
When More Data Does Not Necessarily Mean Better Results
The scientific basis for ORBIL-AI stems from Dr. Oladunni's earlier work on redundancy and complementarity in multimodal machine learning, which he presented at Yale University. His research, including the Complementary Feature Domains (CFD) theory, highlighted that integrating additional feature domains into AI models can introduce overlapping information and increase computational demands without necessarily improving predictive performance. ORBIL-AI expands on this by investigating orthogonality-regularized, information-theoretic learning to develop mathematical and computational methods. This approach enables AI systems to more effectively learn complementary information by studying how computational models represent shared and distinct data across various biomedical observations, potentially improving model generalizability, interpretability, and calibration.
Expanding Biomedical AI Research and Student Opportunities at Morgan
This NIH award significantly strengthens Morgan State University's dedication to fostering investigator-led research, enhancing interdisciplinary collaboration, and creating valuable opportunities for students in emerging scientific fields. The ORBIL-AI project integrates information theory, mathematical representation learning, artificial intelligence, and biomedical signal analysis, thereby contributing to a growing body of research on advanced computational methods for complex scientific challenges. The initiative also builds upon Dr. Oladunni’s existing research in explainable machine learning, multimodal biomedical computing, and theoretical computational representations. Furthermore, it aligns with the NIH SuRE program's objectives to reinforce research capacity and promote meaningful student engagement in biomedical discovery, leveraging Dr. Oladunni's prior visiting faculty research fellowship at Yale University in cardiovascular science and computational AI in medicine.
Advancing the Future of Artificial Intelligence
Although ORBIL-AI focuses on biomedical signals, its core scientific questions about effectively integrating information from multiple sources have broad implications beyond healthcare. As AI systems increasingly depend on diverse data, a crucial challenge is to determine how each source contributes to a computational model’s understanding and whether additional information genuinely enhances its performance. By exploring the mathematical relationships between shared, redundant, and complementary information, ORBIL-AI aims to advance the theoretical foundations of multimodal machine learning. This research holds the potential to inform the development of more efficient, reliable, and interpretable AI systems, driving future innovations in biomedical research and other data-intensive scientific domains. For Morgan State University, this NIH-supported initiative represents a key step in expanding its contributions to foundational artificial intelligence research and preparing the next generation of scientists to tackle evolving technological challenges.
