UT San Antonio Researchers Emphasize Artificial Intelligence

Researchers from the University of Texas at San Antonio (UTSA) shared their findings on the role of artificial intelligence in brain health research at the Neurepiomics Conference and the International Stroke Genetics Consortium annual meeting held in October.

Juan B. Gutiérrez, a mathematics professor at the University of Texas at San Antonio, drew attention to the importance of verifying results generated by artificial intelligence, stating that large language models are mathematically prone to making errors and require independent verification. At the Neurepiomics Conference, Ngoc-Huynh Ho from the Joe R. and Teresa Lozano Long School of Medicine presented research aimed at improving the generalizability of artificial intelligence models used for dementia classification from brain imaging, and addressed the issue of imbalanced training data across different racial and ethnic populations.

In addition, Niyas Shamsudeen Kutty from the Biggs Institute provided information about the Vascular Imaging Biomarker Associations for Alzheimer's Disease and Related Dementias project. This project aims to develop artificial intelligence methods to measure cerebral small vessel disease on MRI datasets obtained from different research centers. The researchers emphasized that while artificial intelligence offers scalability and analytical potential, it requires strict ethical frameworks, transparent methods, and careful evaluation of model performance in order to maintain scientific integrity.