{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:43:26Z","timestamp":1785545006578,"version":"3.56.0"},"reference-count":40,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Distinguishing vitiligo from postinflammatory hypopigmentation (PIH) is clinically challenging because both conditions may present with similar depigmented lesions. Although deep learning has shown strong potential for dermatologic image classification, limited interpretability remains a barrier to clinical adoption.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to develop an interpretable deep learning framework for accurate differentiation between vitiligo and PIH using a lightweight convolutional neural network and an ensemble of explainability methods.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n \n A total of 332 clinical images (176 vitiligo and 156 PIH) were collected from King Abdullah University Hospital and publicly available online sources. Images were preprocessed and evaluated using patient-wise 5-fold cross-validation to eliminate patient-level data leakage. A pretrained MobileNetV2 model was fine-tuned by unfreezing the final 30 layers. To enhance interpretability, gradient-weighted class activation mapping (Grad-CAM), integrated gradients, and smooth gradients (SmoothGrad) were combined into an equal-weight ensemble explanation framework. Performance was assessed using accuracy, precision, recall,\n F<\/jats:italic>\n 1<\/jats:sub>\n -score, and the area under the receiver operating characteristic curve (AUC).\n <\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n The proposed model achieved an overall accuracy of 94.88%, macroaveraged precision of 94.88%, recall of 94.84%,\n F<\/jats:italic>\n 1<\/jats:sub>\n -score of 94.86%, and an AUC of 0.9885 across the 5 validation folds. The ensemble framework produced clinically meaningful explanations in 98.48% of a representative 66-image validation subset used for interpretability analysis.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n The proposed framework combines high diagnostic performance with robust interpretability for distinguishing vitiligo from PIH. By integrating multiple complementary explanation methods, the approach enhances clinical transparency and may support dermatologists in the differential diagnosis of pigmentary disorders.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/81942","type":"journal-article","created":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T19:42:46Z","timestamp":1784922166000},"page":"e81942-e81942","source":"Crossref","is-referenced-by-count":0,"title":["Clinically Interpretable Deep Learning for Differentiating Vitiligo and Postinflammatory Hypopigmentation: Diagnostic Accuracy Study"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3744-2713","authenticated-orcid":false,"given":"Amal Adel","family":"Alzu'bi","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-7335-0347","authenticated-orcid":false,"given":"Shadi","family":"Al Khateeb","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0844-9043","authenticated-orcid":false,"given":"Bassam M","family":"Elzaghmouri","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0244-0221","authenticated-orcid":false,"given":"Diala M","family":"Alshiyab","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2960-8119","authenticated-orcid":false,"given":"Sanaa","family":"Abu Alasal","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4398-0267","authenticated-orcid":false,"given":"Leming","family":"Zhou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,7,24]]},"reference":[{"key":"R1","doi-asserted-by":"publisher","DOI":"10.3389\/fimmu.2023.1309549","article-title":"Large-scale epidemiological analysis of common skin diseases to identify shared and unique comorbidities and demographic factors","volume":"14","author":"Li","journal-title":"Front Immunol"},{"issue":"1","key":"R2","doi-asserted-by":"publisher","first-page":"12","DOI":"10.3390\/make7010012","article-title":"Advancing AI interpretability in medical imaging: a comparative analysis of pixel-level interpretability and Grad-CAM models","volume":"7","author":"Ennab","journal-title":"Mach Learn Knowl Extr"},{"issue":"4","key":"R3","doi-asserted-by":"publisher","first-page":"220","DOI":"10.21015\/vtse.v12i4.2022","article-title":"A review of skin disease detection using deep learning","volume":"12","author":"Fatima","journal-title":"VFAST Trans Softw Eng"},{"issue":"12","key":"R4","doi-asserted-by":"publisher","first-page":"3307","DOI":"10.1007\/s13555-024-01296-9","article-title":"Intelligent diagnosis of hypopigmented dermatoses and intelligent evaluation of vitiligo severity on the basis of deep learning","volume":"14","author":"Huang","journal-title":"Dermatol Ther (Heidelb)"},{"issue":"1","key":"R5","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-024-75062-4","article-title":"AI fusion of multisource data identifies key features of vitiligo","volume":"14","author":"Wang","journal-title":"Sci Rep"},{"key":"R6","doi-asserted-by":"crossref","unstructured":"Griffiths CEM Barker J Bleiker TO Hussain W Simpson RC , editors. Rook\u2019s Textbook of Dermatology. 10th ed. Wiley-Blackwell; 2024. [doi: 10.1002\/9781119709268] ISBN: 9781119709213","DOI":"10.1002\/9781119709268"},{"key":"R7","unstructured":"Kang S Amagai M Bruckner AL , editors. Fitzpatrick\u2019s Dermatology. 9th ed. McGraw-Hill Education; 2019. ISBN: 9780071837798"},{"key":"R8","doi-asserted-by":"crossref","unstructured":"Nijhawan R Mendirtta N Verma M Bohra R Kumar S , editors. Vitiligo detection using machine learning algorithms. In: Intelligent Sustainable Systems. Springer Nature Singapore; 2024:33-42. [doi: 10.1007\/978-981-99-8031-4_4]","DOI":"10.1007\/978-981-99-8031-4_4"},{"issue":"7639","key":"R9","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1038\/nature21056","article-title":"Dermatologist-level classification of skin cancer with deep neural networks","volume":"542","author":"Esteva","journal-title":"Nature"},{"issue":"12","key":"R10","doi-asserted-by":"publisher","first-page":"12623","DOI":"10.1109\/TCYB.2021.3069920","article-title":"Interpretability-based multimodal convolutional neural networks for skin lesion diagnosis","volume":"52","author":"Wang","journal-title":"IEEE Trans Cybern"},{"issue":"8","key":"R11","doi-asserted-by":"publisher","first-page":"1229","DOI":"10.1038\/s41591-020-0942-0","article-title":"Human-computer collaboration for skin cancer recognition","volume":"26","author":"Tschandl","journal-title":"Nat Med"},{"issue":"6","key":"R12","doi-asserted-by":"publisher","first-page":"900","DOI":"10.1038\/s41591-020-0842-3","article-title":"A deep learning system for differential diagnosis of skin diseases","volume":"26","author":"Liu","journal-title":"Nat Med"},{"issue":"1","key":"R13","volume":"8","author":"Usman","journal-title":"J Comput Biomed Inform"},{"key":"R14","doi-asserted-by":"publisher","DOI":"10.1016\/j.cobme.2024.100567","article-title":"Generalizable and explainable deep learning for medical image computing: an overview","volume":"33","author":"Chaddad","journal-title":"Curr Opin Biomed Eng"},{"key":"R15","unstructured":"Dhore V Bhat A Nerlekar V Chavhan K Umare A . Enhancing explainable AI: a hybrid approach combining GradCAM and LRP for CNN interpretability. arXiv. Preprint posted online on May 20, 2024. [doi: 10.48550\/arXiv.2405.12175]"},{"issue":"2","key":"R16","doi-asserted-by":"publisher","first-page":"336","DOI":"10.1007\/s11263-019-01228-7","article-title":"Grad-CAM: visual explanations from deep networks via gradient-based localization","volume":"128","author":"Selvaraju","journal-title":"Int J Comput Vis"},{"key":"R17","unstructured":"Sundararajan M Taly A Yan Q . Axiomatic attribution for deep networks. Presented at: Proceedings of the 34th International Conference on Machine Learning (ICML 2017); Aug 6-11, 2017:3319-3328; Sydney, New South Wales, Australia. URL: https:\/\/proceedings.mlr.press\/v70\/sundararajan17a\/sundararajan17a.pdf [Accessed 07-07-2026]"},{"key":"R18","unstructured":"Smilkov D Thorat N Kim B Vi\u00e9gas F Wattenberg M . SmoothGrad: removing noise by adding noise. arXiv. Preprint posted online on Jun 12, 2017. [doi: 10.48550\/arXiv.1706.03825]"},{"key":"R19","doi-asserted-by":"crossref","unstructured":"Sandler M Howard A Zhu M Zhmoginov A Chen LC . MobileNetV2: inverted residuals and linear bottlenecks. Presented at: 2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition; Jun 18-23, 2018; Salt Lake City, UT, USA. [doi: 10.1109\/CVPR.2018.00474]","DOI":"10.1109\/CVPR.2018.00474"},{"issue":"6","key":"R20","doi-asserted-by":"publisher","DOI":"10.1111\/sji.70076","article-title":"Transforming vitiligo diagnosis and treatment through artificial intelligence: a review","volume":"102","author":"Gholijani","journal-title":"Scand J Immunol"},{"issue":"10","key":"R21","doi-asserted-by":"publisher","DOI":"10.21037\/atm-22-1738","article-title":"A deep learning-based hybrid artificial intelligence model for the detection and severity assessment of vitiligo lesions","volume":"10","author":"Guo","journal-title":"Ann Transl Med"},{"issue":"1","key":"R22","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-024-59436-2","article-title":"Optimizing vitiligo diagnosis with ResNet and Swin transformer deep learning models: a study on performance and interpretability","volume":"14","author":"Zhong","journal-title":"Sci Rep"},{"issue":"3","key":"R23","first-page":"5170","volume":"7","author":"Kantoria","journal-title":"Int Res J Eng Tech"},{"key":"R24","doi-asserted-by":"crossref","unstructured":"Szegedy C Vanhoucke V Ioffe S Shlens J Wojna Z . Rethinking the inception architecture for computer vision. Presented at: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); Jun 27-30, 2016; Las Vegas, NV, USA. [doi: 10.1109\/CVPR.2016.308]","DOI":"10.1109\/CVPR.2016.308"},{"key":"R25","unstructured":"Simonyan K Zisserman A . Very deep convolutional networks for large-scale image recognition. Presented at: 3rd International Conference on Learning Representations, ICLR 2015; May 7-9, 2015; San Diego, CA, USA. URL: https:\/\/bibbase.org\/network\/publication\/simonyan-zisserman-verydeepconvolutionalnetworksforlargescaleimagerecognition-2015 [Accessed 07-07-2026]"},{"key":"R26","unstructured":"Iandola FN Han S Moskewicz MW Ashraf K Dally WJ Keutzer K . SqueezeNet: alexnet-level accuracy with 50x fewer parameters and <0.5MB model size. arXiv. Preprint posted online on Feb 24, 2016. [doi: 10.48550\/arXiv.1602.07360] [Medline: 26778305]"},{"issue":"5","key":"R27","doi-asserted-by":"publisher","first-page":"5479","DOI":"10.1007\/s12652-020-02357-5","article-title":"An effective vitiligo intelligent classification system","volume":"14","author":"Luo","journal-title":"J Ambient Intell Human Comput"},{"key":"R28","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2021.754202","article-title":"Design and assessment of convolutional neural network based methods for vitiligo diagnosis","volume":"8","author":"Zhang","journal-title":"Front Med (Lausanne)"},{"key":"R29","doi-asserted-by":"crossref","unstructured":"Huang G Liu Z Van Der Maaten L Weinberger KQ . Densely connected convolutional networks. Presented at: 2017 IEEE Conference on Computer Vision and Pattern Recognition; Honolulu, HI, USA. [doi: 10.1109\/CVPR.2017.243]","DOI":"10.1109\/CVPR.2017.243"},{"key":"R30","doi-asserted-by":"crossref","unstructured":"Chattopadhay A Sarkar A Howlader P Balasubramanian VN . Grad-cam++: generalized gradient-based visual explanations for deep convolutional networks. Presented at: 2018 IEEE Winter Conference on Applications of Computer Vision (WACV); Mar 12-15, 2018; Lake Tahoe, NV, USA. [doi: 10.1109\/WACV.2018.00097]","DOI":"10.1109\/WACV.2018.00097"},{"key":"R31","doi-asserted-by":"crossref","unstructured":"Wang H Wang Z Du M et al. Score-CAM: score-weighted visual explanations for convolutional neural networks. Presented at: 2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW); Jun 14-19, 2020; Seattle, WA, USA. [doi: 10.1109\/CVPRW50498.2020.00020]","DOI":"10.1109\/CVPRW50498.2020.00020"},{"key":"R32","doi-asserted-by":"publisher","DOI":"10.1177\/20552076251404523","article-title":"Explainable AI for skin disease classification using gradient-weighted class activation mapping and transfer learning in digital health to identify contours","volume":"11","author":"Badhon","journal-title":"Digit Health"},{"key":"R33","doi-asserted-by":"crossref","unstructured":"Liu Z Lin Y Cao Y et al. Swin transformer: hierarchical vision transformer using shifted windows. Presented at: 2021 IEEE\/CVF International Conference on Computer Vision (ICCV); Oct 10-17, 2021; Montreal, QC, Canada. [doi: 10.1109\/ICCV48922.2021.00986]","DOI":"10.1109\/ICCV48922.2021.00986"},{"issue":"4","key":"R34","doi-asserted-by":"publisher","first-page":"1309","DOI":"10.1007\/s00530-021-00787-5","article-title":"An explainable stacked ensemble of deep learning models for improved melanoma skin cancer detection","volume":"28","author":"Shorfuzzaman","journal-title":"Multimed Syst"},{"key":"R35","doi-asserted-by":"crossref","unstructured":"Deng J Dong W Socher R Li LJ Li K Fei-Fei L . ImageNet: a large-scale hierarchical image database. Presented at: 2009 IEEE Conference on Computer Vision and Pattern Recognition; Jun 20-25, 2009; Miami, FL. [doi: 10.1109\/CVPR.2009.5206848]","DOI":"10.1109\/CVPR.2009.5206848"},{"issue":"1","key":"R36","doi-asserted-by":"publisher","DOI":"10.1186\/s12880-022-00793-7","article-title":"Transfer learning for medical image classification: a literature review","volume":"22","author":"Kim","journal-title":"BMC Med Imaging"},{"key":"R37","unstructured":"Molnar C . Interpretable Machine Learning: A Guide for Making Black Box Models Explainable. 3nd ed. Christoph Molnar; 2022. URL: https:\/\/christophm.github.io\/interpretable-ml-book\/ [Accessed 07-07-2026]"},{"key":"R38","unstructured":"Tan M Le QV . EfficientNet: rethinking model scaling for convolutional neural networks. Presented at: the 36th International Conference on Machine Learning; Jun 9-15, 2019; Long Beach, California, USA. URL: https:\/\/dblp.org\/rec\/conf\/icml\/TanL19.html [Accessed 07-07-2026]"},{"key":"R39","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2025.103184","article-title":"Explainability and vision foundation models: a survey","volume":"122","author":"Kazmierczak","journal-title":"Inf Fusion"},{"issue":"1","key":"R40","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-025-57427-z","article-title":"General lightweight framework for vision foundation model supporting multi-task and multi-center medical image analysis","volume":"16","author":"Lu","journal-title":"Nat Commun"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e81942","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T22:26:26Z","timestamp":1785536786000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e81942"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,24]]},"references-count":40,"URL":"https:\/\/doi.org\/10.2196\/81942","relation":{},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,24]]},"article-number":"v14i1e81942"}}