{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T15:12:42Z","timestamp":1785769962595,"version":"3.56.0"},"reference-count":33,"publisher":"MDPI AG","issue":"15","license":[{"start":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T00:00:00Z","timestamp":1785715200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Diagnostics"],"abstract":"<jats:p>Background\/Objectives: Automated dermoscopic image analysis can assist research on early skin cancer detection, but deep learning models may learn acquisition-related shortcuts, including terminal hairs, illumination gradients, gel bubbles, shadows and measurement markings. This study evaluates a causal uncertainty-decomposed ensemble (CUDE) for confounding-aware skin-lesion classification within the ISIC 2019 benchmark setting. Methods: CUDE uses a structured variational autoencoder (SVAE) to impose separate lesion-relevant and acquisition-related nuisance latent heads. Three stochastic latent-space experts are trained on complementary streams and are integrated by a Dirichlet-based fusion module trained with nuisance sampling. The causal graph is used as a modeling prior rather than proof of causal identification. Results: Using a stratified internal split of the labeled ISIC 2019 training collection, CUDE achieved a balanced accuracy of 89.7% and an AUROC of 0.972. Under the predefined synthetic artifact protocol, CUDE showed a relative performance drop of 5.0%, compared with 9.6% for the deep ensemble. Deferring the 10% most uncertain cases increased non-deferred accuracy from 89.7% to 93.2% and reduced false-negative rates for melanoma, BCC and SCC in the non-deferred subset. Conclusions: Within the evaluated ISIC 2019 split and controlled synthetic corruption protocol, structured latent separation, stochastic expert diversity and Dirichlet fusion were associated with improved benchmark robustness and calibration. These results should not be interpreted as evidence of broad real-world robustness or clinical deployment readiness; external multi-center, device-diverse and skin-tone-diverse validation remains required.<\/jats:p>","DOI":"10.3390\/diagnostics16152453","type":"journal-article","created":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T14:16:27Z","timestamp":1785766587000},"page":"2453","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Causal Uncertainty-Decomposed Ensemble Learning for Confounding-Aware Skin Cancer Detection in Dermoscopic Images"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8911-3268","authenticated-orcid":false,"given":"Mohd. Faheem","family":"Khan","sequence":"first","affiliation":[{"name":"Department of Biotechnology and Microbiology, Khandelwal College of Management, Science and Technology, M.J.P. Rohilkhand University, Bareilly 243122, Uttar Pradesh, India"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1095-8445","authenticated-orcid":false,"given":"Khurshid","family":"Ahmad","sequence":"additional","affiliation":[{"name":"Department of Health Informatics, College of Applied Medical Sciences, Qassim University, Buraydah 51452, Saudi Arabia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,3]]},"reference":[{"key":"ref_1","unstructured":"Global Burden of Disease Cancer Collaboration, Fitzmaurice, C., Abate, D., Abbasi, N., Abbastabar, H., Abd-Allah, F., Abdel-Rahman, O., Abdelalim, A., Abdoli, A., and Abdollahpour, I. (2019). Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 29 cancer groups, 1990 to 2017. JAMA Oncol., 5, 1749\u20131768."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Ronchi, A., Argenziano, G., Brancaccio, G., Lallas, A., and Franco, R. (2022). Editorial: Multidisciplinary approach to the diagnosis and therapy of skin neoplasms. Front. Oncol., 12.","DOI":"10.3389\/fonc.2022.939200"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"3022192","DOI":"10.1155\/2024\/3022192","article-title":"Skin cancer segmentation and classification using Vision Transformer for automatic analysis in dermatoscopy-based noninvasive digital system","volume":"2024","author":"Himel","year":"2024","journal-title":"Int. J. Biomed. Imaging"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Tjiu, J., and Lu, C. (2025). Equity and generalizability of artificial intelligence for skin-lesion diagnosis using clinical, dermoscopic, and smartphone images: A systematic review and meta-analysis. Medicina, 61.","DOI":"10.3390\/medicina61122186"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1038\/s41746-020-00380-6","article-title":"Stress testing reveals gaps in clinic readiness of image-based diagnostic artificial intelligence models","volume":"4","author":"Young","year":"2021","journal-title":"npj Digit. Med."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Nauta, M., Walsh, R., Dubowski, A., and Seifert, C. (2021). Uncovering and correcting shortcut learning in machine learning models for skin cancer diagnosis. Diagnostics, 12.","DOI":"10.3390\/diagnostics12010040"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"G\u00f3mez-Mart\u00ednez, V., Seryes-Velasco, C., Chushig-Muzo, D., Godtliebsen, F., Milara, E., and Soguero-Ruiz, C. (2025). Multi-label transformer for dermoscopic artifact detection with an imbalance-aware and confidence-penalizing loss. Res. Sq.","DOI":"10.21203\/rs.3.rs-8057275\/v1"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Bissoto, A., Fornaciali, M., Valle, E., and Avila, S. (2019). (De)Constructing bias on skin lesion datasets. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition Workshops, IEEE.","DOI":"10.1109\/CVPRW.2019.00335"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"107986","DOI":"10.1016\/j.cmpb.2023.107986","article-title":"Dermoscopic dark corner artifacts removal: Friend or foe?","volume":"244","author":"Pewton","year":"2023","journal-title":"Comput. Methods Programs Biomed."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Bissoto, A., Valle, E., and Avila, S. (2020). Debiasing skin lesion datasets and models? Not so fast. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition Workshops, IEEE.","DOI":"10.1109\/CVPRW50498.2020.00378"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Dietterich, T.G. (2000). Ensemble Methods in Machine Learning, Springer.","DOI":"10.1007\/3-540-45014-9_1"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"1290","DOI":"10.1038\/s41598-024-84864-5","article-title":"Boosting skin cancer diagnosis accuracy with ensemble approach","volume":"15","author":"Natha","year":"2025","journal-title":"Sci. Rep."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Shehzad, K., Tan, Z., Shoukat, S., Saeed, A., Ahmad, I., Bhatti, S.S., and Chelloug, S.A. (2023). A deep-ensemble-learning-based approach for skin cancer diagnosis. Electronics, 12.","DOI":"10.3390\/electronics12061342"},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Pearl, J. (2022). Causal diagrams for empirical research. Probabilistic and Causal Inference: The Works of Judea Pearl, Association for Computing Machinery.","DOI":"10.1145\/3501714.3501734"},{"key":"ref_16","unstructured":"Pearl, J., and Mackenzie, D. (2018). The Book of Why: The New Science of Cause and Effect, Basic Books."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"3673","DOI":"10.1038\/s41467-020-17478-w","article-title":"Causality matters in medical imaging","volume":"11","author":"Castro","year":"2020","journal-title":"Nat. Commun."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"457","DOI":"10.1007\/s10994-021-05946-3","article-title":"Aleatoric and epistemic uncertainty in machine learning: An introduction to concepts and methods","volume":"110","author":"Waegeman","year":"2021","journal-title":"Mach. Learn."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1513","DOI":"10.1007\/s10462-023-10562-9","article-title":"A survey of uncertainty in deep neural networks","volume":"56","author":"Gawlikowski","year":"2023","journal-title":"Artif. Intell. Rev."},{"key":"ref_20","unstructured":"Kingma, D.P., and Welling, M. (2013). Auto-encoding variational Bayes. arXiv."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Yang, M., Liu, F., Chen, Z., Shen, X., Hao, J., and Wang, J. (2021). CausalVAE: Disentangled representation learning via neural structural causal models. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR46437.2021.00947"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"103668","DOI":"10.1016\/j.media.2025.103668","article-title":"Robust image representations with counterfactual contrastive learning","volume":"105","author":"Roschewitz","year":"2025","journal-title":"Med. Image Anal."},{"key":"ref_23","unstructured":"Gal, Y., and Ghahramani, Z. (2016, January 19\u201324). Dropout as a Bayesian approximation: Representing model uncertainty in deep learning. Proceedings of the International Conference on Machine Learning, New York, NY, USA."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Neal, R.M. (1996). Bayesian Learning for Neural Networks, Springer.","DOI":"10.1007\/978-1-4612-0745-0"},{"key":"ref_25","unstructured":"\u015eensoy, M., Kaplan, L., and Kandemir, M. (2018, January 3\u20138). Evidential deep learning to quantify classification uncertainty. Proceedings of the 32nd International Conference on Neural Information Processing Systems, Montreal, QC, Canada."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Singh, R.K., Gorantla, R., Allada, S.G., and Narra, P. (2022). SkiNet: A deep learning framework for skin lesion diagnosis with uncertainty estimation and explainability. PLoS ONE, 17.","DOI":"10.1371\/journal.pone.0276836"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Wu, Y., Chen, B., Zeng, A., Pan, D., Wang, R., and Zhao, S. (2022). Skin cancer classification with deep learning: A systematic review. Front. Oncol., 12.","DOI":"10.3389\/fonc.2022.893972"},{"key":"ref_28","first-page":"879","article-title":"3D VAE video prediction model with Kullback Leibler loss enhancement","volume":"21","author":"Dawwd","year":"2024","journal-title":"Int. Arab. J. Inf. Technol."},{"key":"ref_29","unstructured":"Hamilton, W.L., Ying, R., and Leskovec, J. (2017). Inductive representation learning on large graphs. Proceedings of the 31st International Conference on Neural Information Processing Systems, Curran Associates Inc."},{"key":"ref_30","unstructured":"International Skin Imaging Collaboration (2026, June 28). ISIC 2019 Challenge: Skin Lesion Analysis Towards Melanoma Detection. Available online: https:\/\/challenge.isic-archive.com\/landing\/2019\/."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"3211","DOI":"10.1007\/s10994-023-06336-7","article-title":"Classifier calibration: A survey on how to assess and improve predicted class probabilities","volume":"112","author":"Filho","year":"2023","journal-title":"Mach. Learn."},{"key":"ref_32","unstructured":"Tan, M., and Le, Q.V. (2019, January 9\u201315). EfficientNet: Rethinking model scaling for convolutional neural networks. Proceedings of the International Conference on Machine Learning, Long Beach, CA, USA."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Huang, G., Liu, Z., Van der Maaten, L., and Weinberger, K.Q. (2017). Densely connected convolutional networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR.2017.243"}],"container-title":["Diagnostics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/15\/2453\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T14:24:08Z","timestamp":1785767048000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2075-4418\/16\/15\/2453"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,3]]},"references-count":33,"journal-issue":{"issue":"15","published-online":{"date-parts":[[2026,8]]}},"alternative-id":["diagnostics16152453"],"URL":"https:\/\/doi.org\/10.3390\/diagnostics16152453","relation":{},"ISSN":["2075-4418"],"issn-type":[{"value":"2075-4418","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,3]]}}}