Learning with Small Data

Faculty: Pratik Chaudhari

Wang, Rongguang, Pratik Chaudhari, and Christos Davatzikos. “Bias in machine learning models can be significantly mitigated by careful training: Evidence from neuroimaging studies.” Proceedings of the National Academy of Sciences, 2023.

Clinical data is highly heterogeneous.

 

  • e.g., for neurological disorders the heterogeneity stems from diverse anatomies, pathologies, phenotypic and genotypic traits, but also demographic and operational aspects such as data acquisition protocols.

Would like to build diagnostic and prognostic models using multi-source data (imaging, clinical, genetic factors) that can work across broad populations.

 

A representative result where we built diagnostic models of Alzheimer’s disease that can provide unbiased predictions on global-scale data.