npj Digital Medicine (May 2024)

Constructing personalized characterizations of structural brain aberrations in patients with dementia using explainable artificial intelligence

  • Esten H. Leonardsen,
  • Karin Persson,
  • Edvard Grødem,
  • Nicola Dinsdale,
  • Till Schellhorn,
  • James M. Roe,
  • Didac Vidal-Piñeiro,
  • Øystein Sørensen,
  • Tobias Kaufmann,
  • Eric Westman,
  • Andre Marquand,
  • Geir Selbæk,
  • Ole A. Andreassen,
  • Thomas Wolfers,
  • Lars T. Westlye,
  • Yunpeng Wang

DOI
https://doi.org/10.1038/s41746-024-01123-7
Journal volume & issue
Vol. 7, no. 1
pp. 1 – 14

Abstract

Read online

Abstract Deep learning approaches for clinical predictions based on magnetic resonance imaging data have shown great promise as a translational technology for diagnosis and prognosis in neurological disorders, but its clinical impact has been limited. This is partially attributed to the opaqueness of deep learning models, causing insufficient understanding of what underlies their decisions. To overcome this, we trained convolutional neural networks on structural brain scans to differentiate dementia patients from healthy controls, and applied layerwise relevance propagation to procure individual-level explanations of the model predictions. Through extensive validations we demonstrate that deviations recognized by the model corroborate existing knowledge of structural brain aberrations in dementia. By employing the explainable dementia classifier in a longitudinal dataset of patients with mild cognitive impairment, we show that the spatially rich explanations complement the model prediction when forecasting transition to dementia and help characterize the biological manifestation of disease in the individual brain. Overall, our work exemplifies the clinical potential of explainable artificial intelligence in precision medicine.