EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Morphometric Explanations for Deep Learning Based Neuroimaging Classifiers

Ankush Gajanan Arudkar iD, Bernard J E Evans iD, Sabrina Sghirripa iD, Mark Jenkinson iD, Alzheimer's Disease Neuroimaging Initiative

Human Brain Mapping · 2026

DOI: 10.1002/hbm.70633 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Human Brain Mapping, 2026. Verfasst von Ankush Gajanan Arudkar und 4 weiteren Autor:innen. Beteiligte Einrichtung: Adelaide University, Adelaide, Australia. Hauptthemen laut MEDLINE: Deep Learning, Neuroimaging, Magnetic Resonance Imaging, Brain. Die Arbeit nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIA NIH HHS, NIMH NIH HHS, Adelaide University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Despite increasing accuracy of deep‐learning‐based medical imaging classifiers for detecting complex neurodegenerative disorders, an inherent lack of transparency and interpretability in their decisions poses a challenge to their clinical and research applications. While existing explainability methods provide insights into broader aspects of a classification decision, they do not provide a direct interpretation of how specific image information was used for a classification. Geometric, or morphological, brain changes are known to exist in multiple neurodegenerative disorders, and explaining how these changes are being used by a classifier would provide domain experts with an easily interpretable understanding of morphology contributing towards the classification decision. Our work proposes a counterfactual method to explain morphological features used by deep‐learning classifiers for classifying image instances. Furthermore, to provide a more interpretable understanding of common morphological features used by a classifier for the population, we combine the instance explanations to describe average morphological brain changes used by the classifier in MNI152 space. The proposed method was tested using classifiers trained for detecting Parkinson's Disease (PD) and a simulated disease in T1‐w MRIs. The counterfactual images generated by our method were shown to explain learnt disease‐related image information, as the classification of more than 98% of images for the PD classifier was changed by modifying the explained aspects of the image morphology. By deriving geometric explanations used by deep‐learning classifiers, future research can advance trustworthiness by validating the classifiers against existing disease knowledge and potentially harness them to explore novel disease processes. Our code is available at: https://github.com/ankusharudkar/Morphometric‐Explanations .

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningNeuroimagingMagnetic Resonance ImagingBrainParkinson DiseaseImage Processing, Computer-AssistedHumansClassification Algorithms

Stichwörter der Autor:innen

counterfactual explanationexplainable artificial intelligencemagnetic resonance imaging

Autor:innen und Institutionen

  • Adelaide University, Adelaide, Australia.

ORCID-Profile: Ankush Gajanan Arudkar, Bernard J E Evans, Sabrina Sghirripa, Mark Jenkinson

Förderung

  • NIA NIH HHS
  • NIMH NIH HHS
  • Adelaide University
  • NCRR NIH HHS

Zitieren

APA

Arudkar, A. G., Evans, B. J. E., Sghirripa, S., Jenkinson, M., & Alzheimer's Disease Neuroimaging Initiative (2026). Morphometric Explanations for Deep Learning Based Neuroimaging Classifiers. Human Brain Mapping. https://doi.org/10.1002/hbm.70633

Vancouver

Arudkar AG, Evans BJE, Sghirripa S, Jenkinson M, Alzheimer's Disease Neuroimaging Initiative. Morphometric Explanations for Deep Learning Based Neuroimaging Classifiers. Human Brain Mapping. 2026. doi:10.1002/hbm.70633

BibTeX
@article{Arudkar2026,
  author = {Arudkar, Ankush Gajanan and Evans, Bernard J E and Sghirripa, Sabrina and Jenkinson, Mark and Alzheimer's Disease Neuroimaging Initiative, Alzheimer's Disease Neuroimaging Initiative},
  title = {Morphometric Explanations for Deep Learning Based Neuroimaging Classifiers},
  journal = {Human Brain Mapping},
  year = {2026},
  doi = {10.1002/hbm.70633}
}
Kostenlos starten →