Strategies for mitigating data heterogeneities in AI-based neuro-disease detection.

Journal: Neuron
PMID: 40037359

Abstract

In this NeuroView, we discuss challenges and best practices when dealing with disease-detection AI models that are trained on heterogeneous clinical data, focusing on the interrelated problems of model bias, causality, and rare diseases.

Authors

  • Matthew Leming
    Department of Psychiatry, University of Cambridge, Herchel Smith Building for Brain & Mind Sciences, Forvie Site, Robinson Way, Cambridge, CB20SZ, UK.
  • Kyungsu Kim
    Medical AI Research Center, Research Institute for Future Medicine, Samsung Medical Center, Seoul, Republic of Korea. kskim.doc@gmail.com.
  • Rose Bruffaerts
    Laboratory for Cognitive Neurology, Department of Neurosciences, KU Leuven, Herestraat 49, 3000, Leuven, Belgium. rose.bruffaerts@kuleuven.be.
  • Hyungsoon Im
    Center for Systems Biology , Massachusetts General Hospital , Boston , Massachusetts 02114 , United States.

Keywords

External Resources

Popular Topics
Recent Journals