{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T09:22:06Z","timestamp":1787908926372,"version":"build-2784847793"},"reference-count":57,"publisher":"Wiley","issue":"12","license":[{"start":{"date-parts":[[2026,8,27]],"date-time":"2026-08-27T00:00:00Z","timestamp":1787788800000},"content-version":"vor","delay-in-days":12,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,8,15]],"date-time":"2026-08-15T00:00:00Z","timestamp":1786752000000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Human Brain Mapping"],"published-print":{"date-parts":[[2026,8,15]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>\n                    Despite increasing accuracy of deep\u2010learning\u2010based 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\u2010learning 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\u2010w MRIs. The counterfactual images generated by our method were shown to explain learnt disease\u2010related 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\u2010learning 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:\n                    <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" xlink:href=\"https:\/\/github.com\/ankusharudkar\/Morphometric-Explanations\">https:\/\/github.com\/ankusharudkar\/Morphometric\u2010Explanations<\/jats:ext-link>\n                    .\n                  <\/jats:p>","DOI":"10.1002\/hbm.70633","type":"journal-article","created":{"date-parts":[[2026,8,27]],"date-time":"2026-08-27T10:29:38Z","timestamp":1787826578000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Morphometric Explanations for Deep Learning Based Neuroimaging Classifiers"],"prefix":"10.1002","volume":"47","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-3417-7878","authenticated-orcid":false,"given":"Ankush Gajanan","family":"Arudkar","sequence":"first","affiliation":[{"name":"Adelaide University  Adelaide Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3517-3775","authenticated-orcid":false,"given":"Bernard J. E.","family":"Evans","sequence":"additional","affiliation":[{"name":"Adelaide University  Adelaide Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0275-1476","authenticated-orcid":false,"given":"Sabrina","family":"Sghirripa","sequence":"additional","affiliation":[{"name":"Adelaide University  Adelaide Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6043-0166","authenticated-orcid":false,"given":"Mark","family":"Jenkinson","sequence":"additional","affiliation":[{"name":"Adelaide University  Adelaide Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"name":"for the Alzheimer's Disease Neuroimaging Initiative","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,8,27]]},"reference":[{"key":"e_1_2_9_2_1","unstructured":"Agarap A. F.2018.\u201cDeep Learning Using Rectified Linear Units (ReLU).\u201dVersion Number: 2."},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-024-51985-w"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2019.2897538"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.nicl.2018.101645"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2024.3410540"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41531-024-00647-9"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.nicl.2023.103405"},{"key":"e_1_2_9_9_1","unstructured":"Cardoso M. J. W.Li R.Brown et\u00a0al.2022.\u201cMONAI: An Open\u2010Source Framework for Deep Learning in Healthcare.\u201dVersion Number: 1."},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics10060402"},{"key":"e_1_2_9_11_1","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Choi Y.","year":"2018"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1109\/EMBC53108.2024.10782737"},{"key":"e_1_2_9_13_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.retram.2025.103541"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.nicl.2020.102374"},{"key":"e_1_2_9_15_1","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2025.1545409"},{"key":"e_1_2_9_16_1","volume-title":"Advances in Neural Information Processing Systems","author":"Goodfellow I. J.","year":"2014"},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.1007\/11866763_8"},{"key":"e_1_2_9_18_1","first-page":"723","article-title":"A Kernel Two\u2010Sample Test","volume":"13","author":"Gretton A.","year":"2012","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/3236009"},{"key":"e_1_2_9_20_1","unstructured":"He K. X.Zhang S.Ren andJ.Sun.2015.\u201cDeep Residual Learning for Image Recognition.\u201dVersion Number: 1."},{"key":"e_1_2_9_21_1","unstructured":"Heusel M. H.Ramsauer T.Unterthiner B.Nessler andS.Hochreiter.2017.\u201cGANs Trained by a Two Time\u2010Scale Update Rule Converge to a Local Nash Equilibrium.\u201dVersion Number: 6."},{"key":"e_1_2_9_22_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2022.119474"},{"key":"e_1_2_9_23_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-93671-5"},{"key":"e_1_2_9_24_1","doi-asserted-by":"crossref","unstructured":"Isola P. J.\u2010Y.Zhu T.Zhou andA. A.Efros.2017.\u201cImage\u2010To\u2010Image Translation With Conditional Adversarial Networks.\u201d1125\u20131134.","DOI":"10.1109\/CVPR.2017.632"},{"key":"e_1_2_9_25_1","unstructured":"Jaderberg M. K.Simonyan A.Zisserman andK.Kavukcuoglu.2015.\u201cSpatial Transformer Networks.\u201dVersion Number: 3."},{"key":"e_1_2_9_26_1","doi-asserted-by":"publisher","DOI":"10.1006\/nimg.2002.1132"},{"key":"e_1_2_9_27_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2011.09.015"},{"key":"e_1_2_9_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/S1361-8415(01)00036-6"},{"key":"e_1_2_9_29_1","unstructured":"Kingma D. P. andJ.Ba.2014.\u201cAdam: A Method for Stochastic Optimization.\u201dVersion Number: 9."},{"key":"e_1_2_9_30_1","doi-asserted-by":"publisher","DOI":"10.1002\/hbm.23610"},{"key":"e_1_2_9_31_1","doi-asserted-by":"publisher","DOI":"10.3389\/fninf.2013.00045"},{"key":"e_1_2_9_32_1","unstructured":"Lundberg S. andS.\u2010I.Lee.2017.\u201cA Unified Approach to Interpreting Model Predictions.\u201dVersion Number: 2."},{"key":"e_1_2_9_33_1","doi-asserted-by":"publisher","DOI":"10.3389\/fnagi.2024.1498756"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1162\/jocn.2007.19.9.1498"},{"key":"e_1_2_9_35_1","doi-asserted-by":"publisher","DOI":"10.1002\/acn3.644"},{"key":"e_1_2_9_36_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.imu.2024.101587"},{"key":"e_1_2_9_37_1","doi-asserted-by":"publisher","DOI":"10.1212\/WNL.0b013e3181cb3e25"},{"key":"e_1_2_9_38_1","doi-asserted-by":"crossref","unstructured":"Pinaya W. H. L. P.\u2010D.Tudosiu J.Dafflon et\u00a0al.2022.\u201cBrain Imaging Generation With Latent Diffusion Models.\u201dVersion Number: 1.","DOI":"10.1007\/978-3-031-18576-2_12"},{"key":"e_1_2_9_39_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.arr.2016.01.002"},{"key":"e_1_2_9_40_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102723"},{"key":"e_1_2_9_41_1","first-page":"97","volume-title":"Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations","author":"Ribeiro M.","year":"2016"},{"key":"e_1_2_9_42_1","doi-asserted-by":"publisher","DOI":"10.1037\/0033-2909.121.1.133"},{"key":"e_1_2_9_43_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"e_1_2_9_44_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.74"},{"key":"e_1_2_9_45_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102721"},{"key":"e_1_2_9_46_1","unstructured":"Smilkov D. N.Thorat B.Kim F.Vi\u00e9gas andM.Wattenberg.2017.\u201cSmoothGrad: Removing Noise by Adding Noise.\u201dVersion Number: 1."},{"key":"e_1_2_9_47_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2024.3412216"},{"key":"e_1_2_9_48_1","doi-asserted-by":"publisher","DOI":"10.1109\/JBHI.2022.3172976"},{"key":"e_1_2_9_49_1","unstructured":"Sundararajan M. A.Taly andQ.Yan.2017.\u201cAxiomatic Attribution for Deep Networks.\u201dVersion Number: 2."},{"key":"e_1_2_9_50_1","doi-asserted-by":"publisher","DOI":"10.1111\/jnc.13691"},{"key":"e_1_2_9_51_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00429-009-0203-y"},{"key":"e_1_2_9_52_1","doi-asserted-by":"crossref","unstructured":"Tran D. L.Bourdev R.Fergus L.Torresani andM.Paluri.2014.\u201cLearning Spatiotemporal Features With 3D Convolutional Networks.\u201dVersion Number: 4.","DOI":"10.1109\/ICCV.2015.510"},{"key":"e_1_2_9_53_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-024-00864-0"},{"key":"e_1_2_9_54_1","unstructured":"Ulyanov D. A.Vedaldi andV.Lempitsky.2016.\u201cInstance Normalization: The Missing Ingredient for Fast Stylization.\u201dVersion Number: 3."},{"key":"e_1_2_9_55_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2025.110078"},{"key":"e_1_2_9_56_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102470"},{"key":"e_1_2_9_57_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACSSC.2003.1292216"},{"key":"e_1_2_9_58_1","doi-asserted-by":"publisher","DOI":"10.1109\/42.906424"}],"container-title":["Human Brain Mapping"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/hbm.70633","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/hbm.70633","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/hbm.70633","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T09:13:22Z","timestamp":1787908402000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/hbm.70633"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,15]]},"references-count":57,"journal-issue":{"issue":"12","published-print":{"date-parts":[[2026,8,15]]}},"alternative-id":["10.1002\/hbm.70633"],"URL":"https:\/\/doi.org\/10.1002\/hbm.70633","archive":["Portico"],"relation":{},"ISSN":["1065-9471","1097-0193"],"issn-type":[{"value":"1065-9471","type":"print"},{"value":"1097-0193","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,15]]},"assertion":[{"value":"2026-01-26","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-08-16","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-08-27","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70633"}}