EPSCoR Research Fellows: NSF: Interpretable AI-Driven Multi-Modal MRI for Brain Imaging Analysis

This Research Infrastructure Improvement (RII) EPSCoR Research Fellows project provides a fellowship to an Assistant professor and training for a graduate student at the Tulane University. This work is conducted in collaboration with experts at the Georgia State University/Georgia Institute of Technology/Emory University Center for Translational Research in Neuroimaging and Data Science (TReNDS). Through the fellowship, the PI will investigate how artificial intelligence can analyze complex brain imaging data while providing explanations that are transparent, reliable, and consistent with expert reasoning. The project integrates computer science and neuroscience to design AI systems that analyze brain scans and explain their results. It addresses the challenge of making AI transparent and trustworthy for medical and research applications. The project will help build a skilled STEM workforce while promoting the development of trustworthy AI technologies that can be applied in healthcare and other high-stakes fields, including education, public policy, and industry.

The project will develop interpretable artificial intelligence models for multi-modal brain imaging to improve understanding of brain function and support clinical decision-making. It will advance both Artificial Intelligence (AI) and neuroscience by creating models that generate human-understandable explanations from complex imaging data, allowing researchers and clinicians to interpret, validate, and trust predictions. The project will design inherently interpretable AI architectures for structural, functional, and diffusion MRI, and implement algorithms for transparent multi-modal data integration, providing a holistic view of brain structure and activity. Models will incorporate expert knowledge and neuroscientific principles, and outputs will be evaluated with domain experts to ensure scientific and clinical relevance. The project will support the PI’s professional development in brain imaging and provide PhD students with training in AI and neuroimaging. Activities will integrate curriculum development, student mentoring, and interdisciplinary research at the home institution to strengthen STEM capacity and workforce development.

This project is supported by the EPSCoR Research Infrastructure Improvement Program: EPSCoR Research Fellows (ERF). The ERF program supports early- and mid-career investigators in eligible jurisdictions to develop collaborations at the nation’s private, government or academic research institutions.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Award Number: 2531809
Principal Investigator: Zhengming Ding
Funds Obligated: $222,497
State: LA