Federal Grant Award Summary

The University of South Carolina received a $1,016,319 Project Grant award from the National Institute of Mental Health (NIMH) under the Mental Health Research Grants program (CFDA 93.242), effective July 1, 2026, with completion targeted for April 30, 2031. The project, titled "Multimodal AI for Monitoring and Predicting Neurocognitive Impairment in People with HIV," will develop and apply artificial intelligence methodologies to advance precision medicine approaches for HIV-associated neurocognitive disorders (HAND). The research will leverage the All of Us Research Program, a centralized NIH data repository providing secure access to de-identified multimodal data including electronic health records, genomic information, survey responses, and imaging data from nearly one million participants, including over 6,600 individuals with confirmed HIV infection identified through computational phenotyping.

The grant deliverables will focus on creating integrated machine learning and deep learning models that combine multiple data modalities to improve the diagnostic accuracy and prognostic capability of existing neurocognitive impairment assessment tools. Current clinical approaches remain limited by siloed, unimodal datasets; this project will address that gap by developing multimodal artificial intelligence methodologies that leverage complementary information across data types to enhance real-time monitoring and clinical decision-making for the nearly half of people with HIV who experience asymptomatic neurocognitive disorder or mild neurocognitive disorder despite advances in antiretroviral therapy.