Forfatter
Tittel Constructing personalized characterizations of structural brain aberrations in patients with dementia using explainable artificial intelligence
Medansvarlig Persson, Karin Grødem, Edvard Dinsdale, Nicola Schellhorn, Till Roe, James M Vidal-Piñeiro, Didac Sørensen, Øystein Kaufmann, Tobias Westman, Eric Marquand, Andre F Selbæk, Geir Andreassen, Ole A Wolfers, Thomas Westlye, Lars T. Wang, Yunpeng
Språk
Emne
Serie
Serienummer vol 7, Article number: 110
År
Noter 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.
Tilgjengelige
L�nere p� venteliste
*000 ap
*00142175
*100 $aLeonardsen, Esten H.
*245 $aConstructing personalized characterizations of structural brain aberrations in patients with dementia using explainable artificial intelligence$cEsten 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
*260 $c2024
*300 $ahttps://doi.org/10.1038/s41746-024-01123-7
*440 $anpj Digital Medicine $vvol 7, Article number: 110
*505 $aAbstract: 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.
*650 $aKunstig intelligens
*650 $aDemens
*700 $aPersson, Karin
*700 $aGrødem, Edvard
*700 $aDinsdale, Nicola
*700 $aSchellhorn, Till
*700 $aRoe, James M
*700 $aVidal-Piñeiro, Didac
*700 $aSørensen, Øystein
*700 $aKaufmann, Tobias
*700 $aWestman, Eric
*700 $aMarquand, Andre F
*700 $aSelbæk, Geir
*700 $aAndreassen, Ole A
*700 $aWolfers, Thomas
*700 $aWestlye, Lars T.
*700 $aWang, Yunpeng
*856 $uhttps://doi.org/10.1038/s41746-024-01123-7
^
Det finnes ingen anmeldelser for denne boken.
Klikk her
for bli den første til å gi din mening