{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,29]],"date-time":"2026-06-29T14:27:54Z","timestamp":1782743274952,"version":"3.54.5"},"reference-count":24,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,12,24]],"date-time":"2025-12-24T00:00:00Z","timestamp":1766534400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,12,24]],"date-time":"2025-12-24T00:00:00Z","timestamp":1766534400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Health Inf Sci Syst"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Deep learning based diagnostic AI systems based on medical images are starting to provide similar performance as human experts. However, these data-hungry complex systems are inherently black boxes and therefore slow to be adopted for high-risk applications like healthcare. This problem of lack of transparency is exacerbated in the case of recent large foundation models, which are trained in a self-supervised manner on millions of data points to provide robust generalisation across a range of downstream tasks. The embeddings generated from them happen through a process that is not interpretable, and hence not easily trustable for clinical applications. To address this timely issue, we deploy conformal analysis to quantify the predictive uncertainty of a vision transformer (ViT)-based foundation model across patient demographics with respect to sex, age, and ethnicity for the task of skin lesion classification using several public benchmark datasets. The significant advantage of this method is that conformal analysis is method independent, and it not only provides a coverage guarantee at the population level but also provides an uncertainty score for each individual. This is used to demonstrate the effectiveness of utilizing these embeddings for specialized tasks like diagnostic classification, meanwhile reducing computational costs. Secondly, the public benchmark datasets we used had severe class imbalance in terms of the number of samples in different classes. We used a model-agnostic dynamic F1-score-based sampling during model training, which helped to stabilize the class imbalance. We investigate the effects on uncertainty quantification (UQ) with or without this bias mitigation step. Thus, our results show how this can be used as a fairness metric to evaluate the robustness of the feature embeddings of the foundation model (Google DermFoundation), advancing the trustworthiness and fairness of clinical AI.<\/jats:p>","DOI":"10.1007\/s13755-025-00412-z","type":"journal-article","created":{"date-parts":[[2025,12,24]],"date-time":"2025-12-24T18:12:16Z","timestamp":1766599936000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Conformal uncertainty quantification to evaluate predictive fairness of foundation AI model for skin lesion classes across patient demographics"],"prefix":"10.1007","volume":"14","author":[{"given":"Swarnava","family":"Bhattacharyya","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Umapada","family":"Pal","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5597-908X","authenticated-orcid":false,"given":"Tapabrata","family":"Chakraborti","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,12,24]]},"reference":[{"issue":"12","key":"412_CR1","doi-asserted-by":"publisher","first-page":"1749","DOI":"10.1001\/jamaoncol.2019.2996","volume":"5","author":"C Fitzmaurice","year":"2019","unstructured":"Fitzmaurice C. 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 a systematic analysis for the global burden of disease study. JAMA Oncol. 2019;5(12):1749\u201368. https:\/\/doi.org\/10.1001\/jamaoncol.2019.2996.","journal-title":"JAMA Oncol"},{"key":"412_CR2","unstructured":"American Cancer Society. Cancer Facts and Figures, (2025), p. 22. https:\/\/www.cancer.org\/content\/dam\/cancer-org\/research\/cancer-facts-and-statistics\/annual-cancer-facts-and-figures\/2025\/2025-cancer-facts-and-figures-acs.pdf"},{"key":"412_CR3","unstructured":"Skin lesion detection using statistical features and traditional machine learning methods: A review. Kaur K, Sharma D, Kumar A, Kaur P, Gencoglan DN. Integrated Technologies in Electrical, Electronics and Biotechnology Engineering."},{"issue":"1","key":"412_CR4","doi-asserted-by":"publisher","first-page":"e3212204","DOI":"10.17061\/phrp3212204","volume":"32","author":"M Janda","year":"2022","unstructured":"Janda M, Olsen CM, Mar VJ, Cust AE. Early detection of skin cancer in Australia\u2014current approaches and new opportunities. Public Health Res Pract. 2022;32(1):e3212204. https:\/\/doi.org\/10.17061\/phrp3212204.","journal-title":"Public Health Res Pract"},{"key":"412_CR5","doi-asserted-by":"publisher","first-page":"1183","DOI":"10.3390\/healthcare10071183","volume":"10","author":"W Gouda","year":"2022","unstructured":"Gouda W, Sama NU, Al-Waakid G, Humayun M, Jhanjhi NZ. Detection of skin cancer based on skin lesion images using deep learning. Healthcare. 2022;10:1183. https:\/\/doi.org\/10.3390\/healthcare10071183.","journal-title":"Healthcare"},{"key":"412_CR6","doi-asserted-by":"publisher","unstructured":"Garcia SI. Meta-learning for skin cancer detection using deep learning techniques. https:\/\/doi.org\/10.48550\/arXiv.2104.10775","DOI":"10.48550\/arXiv.2104.10775"},{"key":"412_CR7","doi-asserted-by":"publisher","unstructured":"Zhang Y, Gao J, Zhou M, Wang X, et al. Text-guided foundation model adaptation for pathological image classification. MICCAI 2023. Lecture Notes in Computer Science, vol 14224. https:\/\/doi.org\/10.48550\/arXiv.2307.14901","DOI":"10.48550\/arXiv.2307.14901"},{"key":"412_CR8","unstructured":"Khoiwal RH, McMillan AB. Embeddings are all you need! achieving high performance medical image classification through training-free embedding analysis. https:\/\/arxiv.org\/abs\/2412.09445"},{"key":"412_CR9","unstructured":"Restrepo D, Wu C, Cajas SA, Nakayama LF, et al. Multimodal deep learning for low-resource settings: a vector embedding alignment approach for healthcare applications. https:\/\/arxiv.org\/abs\/2406.02601"},{"key":"412_CR10","unstructured":"Codella NCF, Jin Y, Jain S, Gu Y, et al. MedImageInsight: an open-source embedding model for general domain medical imaging. https:\/\/arxiv.org\/abs\/2410.06542"},{"key":"412_CR11","doi-asserted-by":"publisher","first-page":"273","DOI":"10.1007\/s10462-024-10884-2","volume":"57","author":"M Salmi","year":"2024","unstructured":"Salmi M, Atif D, Oliva D, Abraham A, Ventura S. Handling imbalanced medical datasets: review of a decade of research. Artif Intell Rev. 2024;57:273. https:\/\/doi.org\/10.1007\/s10462-024-10884-2.","journal-title":"Artif Intell Rev"},{"key":"412_CR12","doi-asserted-by":"publisher","first-page":"2897","DOI":"10.3390\/cancers14122897","volume":"14","author":"E Tasci","year":"2022","unstructured":"Tasci E, Zhuge Y, Camphausen K, Krauze AV. Bias and class imbalance in oncologic data-towards inclusive and transferrable AI in large scale oncology data sets. Cancers. 2022;14:2897. https:\/\/doi.org\/10.3390\/cancers14122897.","journal-title":"Cancers"},{"key":"412_CR13","unstructured":"Kolesnikov A, Beyer L, Zhai X, Puigcerver J, Yung J, Gelly S, Houlsby N. Big transfer (BiT): general visual representation learning. https:\/\/arxiv.org\/abs\/1912.11370"},{"key":"412_CR14","doi-asserted-by":"publisher","first-page":"7761","DOI":"10.1038\/s41467-022-34945-8","volume":"13","author":"H Olsson","year":"2022","unstructured":"Olsson H, Kartasalo K, Mulliqi N, Capuccini M, et al. Estimating diagnostic uncertainty in artificial intelligence assisted pathology using conformal prediction. Nat Commun. 2022;13:7761. https:\/\/doi.org\/10.1038\/s41467-022-34945-8.","journal-title":"Nat Commun"},{"key":"412_CR15","unstructured":"Hui L. Risk-sensitive conformal prediction for catheter placement detection in chest X-rays. https:\/\/arxiv.org\/abs\/2505.22496"},{"key":"412_CR16","doi-asserted-by":"publisher","first-page":"26697","DOI":"10.1038\/s41598-025-09580-0","volume":"15","author":"Z Shen","year":"2025","unstructured":"Shen Z, Chakraborti T, Banerji CRS, Ding X. Conformal prediction quantifies wearable cuffless blood pressure with certainty. Sci Rep. 2025;15:26697. https:\/\/doi.org\/10.1038\/s41598-025-09580-0.","journal-title":"Sci Rep"},{"key":"412_CR17","doi-asserted-by":"publisher","unstructured":"Shen Z, Chakraborti T, Wang W, Yao S, et al. Uncertainty quantification of Cuffless blood pressure estimation based on parameterized model evidential ensemble learning. https:\/\/doi.org\/10.1016\/j.bspc.2024.106104","DOI":"10.1016\/j.bspc.2024.106104"},{"key":"412_CR18","unstructured":"Dey S, Basuchowdhuri P, Mitra D, et al. Uncertainty estimation using boundary prediction for super-resolved images. https:\/\/www.sciencedirect.com\/science\/article\/abs\/pii\/S1077314225000724"},{"key":"412_CR19","doi-asserted-by":"publisher","unstructured":"Dey S, Banerji CRS, Basuchowdhuri P, et al. Generating crossmodal gene expression from cancer histopathology improves multimodal AI predictions. https:\/\/doi.org\/10.48550\/arXiv.2502.00568","DOI":"10.48550\/arXiv.2502.00568"},{"key":"412_CR20","doi-asserted-by":"publisher","unstructured":"Chakraborti T, Banerji CRS, Marandon A, Hellon V, Mitra R, et al. Personalized uncertainty quantification in artificial intelligence. https:\/\/doi.org\/10.1038\/s42256-025-01024-8","DOI":"10.1038\/s42256-025-01024-8"},{"issue":"3","key":"412_CR21","first-page":"371","volume":"9","author":"G Shafer","year":"2008","unstructured":"Shafer G, Vovk V. A tutorial on conformal prediction. J Mach Learning Res. 2008;9(3):371\u2013421.","journal-title":"J Mach Learning Res"},{"key":"412_CR22","doi-asserted-by":"publisher","unstructured":"Angelopoulos AN, Barber RF, Bates S. Theoretical foundations of conformal prediction. https:\/\/doi.org\/10.48550\/arXiv.2411.11824","DOI":"10.48550\/arXiv.2411.11824"},{"key":"412_CR23","doi-asserted-by":"publisher","unstructured":"Combalia M, Codella NCF, Rotemberg V, et al. BCN20000: Dermascopic Lesions in the Wild. https:\/\/doi.org\/10.48550\/arXiv.1908.02288","DOI":"10.48550\/arXiv.1908.02288"},{"issue":"7","key":"412_CR24","doi-asserted-by":"publisher","first-page":"1529","DOI":"10.1016\/j.jid.2018.01.028","volume":"138","author":"SS Han","year":"2018","unstructured":"Han SS, Kim MS, Lim W, Park GH, Park I, Chang SE. Classification of the clinical images for benign and malignant cutaneous tumors using a deep learning algorithm. J Investig Dermatol. 2018;138(7):1529\u201338. https:\/\/doi.org\/10.1016\/j.jid.2018.01.028.","journal-title":"J Investig Dermatol"}],"container-title":["Health Information Science and Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13755-025-00412-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s13755-025-00412-z","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13755-025-00412-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,24]],"date-time":"2025-12-24T18:12:18Z","timestamp":1766599938000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s13755-025-00412-z"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,24]]},"references-count":24,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["412"],"URL":"https:\/\/doi.org\/10.1007\/s13755-025-00412-z","relation":{},"ISSN":["2047-2501"],"issn-type":[{"value":"2047-2501","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,24]]},"assertion":[{"value":"11 March 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"9 December 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"24 December 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Authors declare no Conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}],"article-number":"18"}}