{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,7,2]],"date-time":"2025-07-02T04:14:37Z","timestamp":1751429677709,"version":"3.41.0"},"reference-count":31,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,5,16]],"date-time":"2025-05-16T00:00:00Z","timestamp":1747353600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,5,16]],"date-time":"2025-05-16T00:00:00Z","timestamp":1747353600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100005760","name":"University of Gothenburg","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100005760","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BMC Med Imaging"],"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:sec>\n            <jats:title>Background<\/jats:title>\n            <jats:p>Basal cell carcinoma (BCC) is the most common skin cancer, placing a significant burden on healthcare systems globally. Developing high-precision automated diagnostics requires large annotated datasets, which are costly and difficult to obtain. This study aimed to fine-tune a weakly supervised machine learning model to classify BCC in preoperative punch biopsies using transfer learning. By addressing challenges of scalability and variability, this approach seeks to enhance generalizability and diagnostic accuracy.<\/jats:p>\n          <\/jats:sec>\n          <jats:sec>\n            <jats:title>Methods<\/jats:title>\n            <jats:p>The Basal Cell Classification (BCCC) dataset included 514 WSIs of punch biopsies (261 with BCC and 253 tumor-free slides), divided into training (70%), validation (15%), and test sets (15%). WSIs were split into patches, and features were extracted using a pretrained simCLR model trained on 1,435 WSIs from BCC excisions. Features were formed into graphs for spatial information and the processed by a Vision Transformer. Testing included finetuned and non-finetuned pre-trained models as well as a model trained from the scratch, evaluated on 78 WSIs from the BCCC dataset. The COBRA dataset of 3,588 WSIs (1,794 with BCC and 1,794 without) was used for external validation. Models classified no-tumor vs. tumor (two classes), no-tumor vs. low-risk vs. high-risk tumors (three classes), and no-tumor vs. four BCC subtypes (five classes).<\/jats:p>\n          <\/jats:sec>\n          <jats:sec>\n            <jats:title>Results<\/jats:title>\n            <jats:p>The fine-tuned model significantly outperformed the non-fine-tuned pretrained model and the model trained from the scratch with accuracies of 91.7%, 82.1%, and 75.3% and with AUCs of 0.98, 0.95\u20130.98, and 0.91\u20130.97 for two, three, and five-class classification. On the external validation, accuracies were 84.9% and 70.5%, with AUCs of 0.92 and 0.89\u20130.91 for two and three-class classification, respectively. The ablation study revealed that the fine-tuned model outperformed the model trained from scratch, improving mean accuracy by 10.6%, 11.7%, and 13.1% on the BCCC dataset, as well as by 29.6% and 19.2% on the COBRA dataset.<\/jats:p>\n          <\/jats:sec>\n          <jats:sec>\n            <jats:title>Conclusions<\/jats:title>\n            <jats:p>The results suggest that transfer learning not only enhances model performance on small datasets but also supports robust feature extraction in complex histopathology tasks. These findings reinforce the utility of pre-trained models in computational pathology, where access to large, labeled datasets is often limited, and task-specific challenges require nuanced understanding of the visual data.<\/jats:p>\n          <\/jats:sec>","DOI":"10.1186\/s12880-025-01710-4","type":"journal-article","created":{"date-parts":[[2025,5,16]],"date-time":"2025-05-16T11:18:55Z","timestamp":1747394335000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Enhancing basal cell carcinoma classification in preoperative biopsies via transfer learning with weakly supervised graph transformers"],"prefix":"10.1186","volume":"25","author":[{"given":"Johan","family":"Bj\u00f6rkman","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sigrid","family":"Lagerroth","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jan","family":"Siarov","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Filmon","family":"Yacob","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Noora","family":"Neittaanm\u00e4ki","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2025,5,16]]},"reference":[{"key":"1710_CR1","doi-asserted-by":"publisher","first-page":"356","DOI":"10.1016\/j.ejca.2018.07.005","volume":"103","author":"J Ferlay","year":"2018","unstructured":"Ferlay J, Colombet M, Soerjomataram I, Dyba T, Randi G, Bettio M, et al. Cancer incidence and mortality patterns in Europe: estimates for 40 countries and 25 major cancers in 2018. Eur J Cancer. 2018;103:356\u201387.","journal-title":"Eur J Cancer"},{"key":"1710_CR2","doi-asserted-by":"publisher","first-page":"367","DOI":"10.1111\/ced.12016","volume":"38","author":"NJ Levell","year":"2013","unstructured":"Levell NJ, Igali L, Wright KA, Greenberg DC. Basal cell carcinoma epidemiology in the UK: the elephant in the room. Clin Exp Dermatol. 2013;38:367\u20139.","journal-title":"Clin Exp Dermatol"},{"key":"1710_CR3","doi-asserted-by":"publisher","first-page":"837","DOI":"10.1001\/archderm.140.7.837","volume":"140","author":"KG Lewis","year":"2004","unstructured":"Lewis KG, Weinstock MA. Nonmelanoma skin cancer mortality (1988\u20132000): the Rhode Island follow-back study. Arch Dermatol. 2004;140:837\u201342.","journal-title":"Arch Dermatol"},{"key":"1710_CR4","first-page":"167","volume":"88","author":"AG Marzuka","year":"2015","unstructured":"Marzuka AG, Book SE. Basal cell carcinoma: pathogenesis, epidemiology, clinical features, diagnosis, histopathology, and management. Yale J Biol Med. 2015;88:167\u201379.","journal-title":"Yale J Biol Med"},{"key":"1710_CR5","doi-asserted-by":"publisher","first-page":"540","DOI":"10.1016\/j.jaad.2017.10.006","volume":"78","author":"JYS Kim","year":"2018","unstructured":"Kim JYS, Kozlow JH, Mittal B, Moyer J, Olencki T, Rodgers P. Guidelines of care for the management of basal cell carcinoma. J Am Acad Dermatol. 2018;78:540\u201359.","journal-title":"J Am Acad Dermatol"},{"key":"1710_CR6","volume-title":"WHO classification of skin tumours","author":"DE Elder","year":"2018","unstructured":"Elder DE, Scolyer RA, Willemze R. WHO classification of skin tumours. 4th ed. Berlin: WHO; 2018.","edition":"4"},{"key":"1710_CR7","first-page":"3467","volume":"85","author":"J Jernbeck","year":"1988","unstructured":"Jernbeck J, Glaumann B, Glas JE. Clinical evaluation of the histological grading of aggressive types of cancer. Lakartidningen. 1988;85:3467\u201370.","journal-title":"Lakartidningen"},{"issue":"4","key":"1710_CR8","first-page":"542","volume":"32","author":"PR Cohen","year":"2006","unstructured":"Cohen PR, Schulze KE, Nelson BR. Basal cell carcinoma with mixed histology: a possible pathogenesis for recurrent skin cancer. Dermatol Surg. 2006;32(4):542\u201351.","journal-title":"Dermatol Surg"},{"key":"1710_CR9","doi-asserted-by":"publisher","first-page":"279","DOI":"10.1016\/j.jaad.2007.03.021","volume":"57","author":"J Jagdeo","year":"2007","unstructured":"Jagdeo J, Weinstock MA, Piepkorn M, Bingham SF, Department of Veteran Affairs Topical Tretinoin Chemoprevention Trial Group. Reliability of the histopathologic diagnosis of keratinocyte carcinomas. J Am Acad Dermatol. 2007;57:279\u201384.","journal-title":"J Am Acad Dermatol"},{"key":"1710_CR10","doi-asserted-by":"publisher","first-page":"854","DOI":"10.1001\/jamadermatol.2019.0066","volume":"155","author":"DJ Moon","year":"2019","unstructured":"Moon DJ, Higgins S, Feinstein S, Ahadiat O, Sutton A, Wysong A. Variance of basal cell carcinoma subtype reporting by practice setting. JAMA Dermatol. 2019;155:854\u20136.","journal-title":"JAMA Dermatol"},{"key":"1710_CR11","doi-asserted-by":"publisher","first-page":"141","DOI":"10.1097\/CEJ.0000000000000056","volume":"24","author":"LG Gordon","year":"2015","unstructured":"Gordon LG, Rowell D. Health system costs of skin cancer and cost-effectiveness of skin cancer prevention and screening: a systematic review. Eur J Cancer Prev. 2015;24:141\u20139.","journal-title":"Eur J Cancer Prev"},{"key":"1710_CR12","doi-asserted-by":"publisher","first-page":"62","DOI":"10.1111\/joim.13030","volume":"288","author":"B Acs","year":"2020","unstructured":"Acs B, Rantalainen M, Hartman J. Artificial intelligence as the next step towards precision pathology. J Intern Med. 2020;288:62\u201381.","journal-title":"J Intern Med"},{"key":"1710_CR13","doi-asserted-by":"publisher","first-page":"100357","DOI":"10.1016\/j.jpi.2023.100357","volume":"15","author":"MS Hosseini","year":"2024","unstructured":"Hosseini MS, Bejnordi BE, Trinh VQ, Chan L, Hasan D, Li X, et al. Computational pathology: A survey review and the way forward. J Pathol Inf. 2024;15:100357.","journal-title":"J Pathol Inf"},{"key":"1710_CR14","doi-asserted-by":"publisher","first-page":"32","DOI":"10.4103\/jpi.jpi_31_18","volume":"9","author":"TG Olsen","year":"2018","unstructured":"Olsen TG, Jackson BH, Feeser TA, Kent MN, Moad JC, Krishnamurthy S, et al. Diagnostic performance of deep learning algorithms applied to three common diagnoses in dermatopathology. J Pathol Inf. 2018;9:32.","journal-title":"J Pathol Inf"},{"issue":"1","key":"1710_CR15","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1093\/ajcp\/aqab085","volume":"157","author":"ED McAlpine","year":"2022","unstructured":"McAlpine ED, Michelow P, Liebenberg E, Pantanowitz L. The utility of unsupervised machine learning in anatomic pathology. Am J Clin Pathol. 2022;157(1):5\u201314.","journal-title":"Am J Clin Pathol"},{"key":"1710_CR16","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1038\/nature14539","volume":"521","author":"Y LeCun","year":"2015","unstructured":"LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015;521:436\u201344.","journal-title":"Nature"},{"issue":"8","key":"1710_CR17","doi-asserted-by":"publisher","first-page":"1301","DOI":"10.1038\/s41591-019-0508-1","volume":"25","author":"G Campanella","year":"2019","unstructured":"Campanella G, Hanna MG, Geneslaw L, Miraflor A, Werneck Krauss Silva V, Busam KJ, et al. Clinical-grade computational pathology using weakly supervised deep learning on whole slide images. Nat Med. 2019;25(8):1301\u20139.","journal-title":"Nat Med"},{"issue":"1","key":"1710_CR18","doi-asserted-by":"publisher","first-page":"7555","DOI":"10.1038\/s41598-023-33863-z","volume":"13","author":"F Yacob","year":"2023","unstructured":"Yacob F, Siarov J, Villiamsson K, Suvilehto JT, Sj\u00f6blom L, Kjellberg M, et al. Weakly supervised detection and classification of basal cell carcinoma using graph-transformer on whole slide images. Sci Rep. 2023;13(1):7555.","journal-title":"Sci Rep"},{"key":"1710_CR19","doi-asserted-by":"publisher","first-page":"100335","DOI":"10.1016\/j.jpi.2023.100335","volume":"14","author":"K Al-Thelaya","year":"2023","unstructured":"Al-Thelaya K, Gilal NU, Alzubaidi M, Majeed F, Agus M, Schneider J, et al. Applications of discriminative and deep learning feature extraction methods for whole slide image analysis: A survey. J Pathol Inf. 2023;14:100335.","journal-title":"J Pathol Inf"},{"key":"1710_CR20","doi-asserted-by":"publisher","first-page":"131","DOI":"10.1007\/s00428-019-02594-w","volume":"475","author":"D Komura","year":"2019","unstructured":"Komura D, Ishikawa S. Machine learning approaches for pathologic diagnosis. Virchows Arch. 2019;475:131\u20138.","journal-title":"Virchows Arch"},{"key":"1710_CR21","doi-asserted-by":"publisher","first-page":"102027","DOI":"10.1016\/j.compmedimag.2021.102027","volume":"95","author":"D Ahmedt-Aristizabal","year":"2022","unstructured":"Ahmedt-Aristizabal D, Armin MA, Denman S, Fookes C, Petersson L. A survey on graph-based deep learning for computational histopathology. Comput Med Imaging Graph. 2022;95:102027.","journal-title":"Comput Med Imaging Graph"},{"key":"1710_CR22","unstructured":"Dosovitskiy A, Beyer L, Kolesnikov A, Weissenborn D, Zhai X, Unterthiner T et al. An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. arXiv:2010.11929 [Preprint]. 2020 [cited 2024 Aug 28]. Available from: https:\/\/arxiv.org\/abs\/2010.11929"},{"issue":"1","key":"1710_CR23","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1186\/s12938-023-01157-0","volume":"22","author":"CC Atabansi","year":"2023","unstructured":"Atabansi CC, Nie J, Liu H, Song Q, Yan L, Zhou X. A survey of transformer applications for histopathological image analysis: new developments and future directions. Biomed Eng Online. 2023;22(1):96.","journal-title":"Biomed Eng Online"},{"key":"1710_CR24","doi-asserted-by":"publisher","first-page":"1099","DOI":"10.1136\/amiajnl-2012-001540","volume":"20","author":"S Kothari","year":"2013","unstructured":"Kothari S, Phan JH, Stokes TH, Wang MD. Pathology imaging informatics for quantitative analysis of whole-slide images. J Am Med Inf Assoc. 2013;20:1099\u2013108.","journal-title":"J Am Med Inf Assoc"},{"issue":"12","key":"1710_CR25","doi-asserted-by":"publisher","first-page":"1420","DOI":"10.1038\/s41551-022-00929-8","volume":"6","author":"C Chen","year":"2022","unstructured":"Chen C, Lu MY, Williamson DFK, Chen TY, Schaumberg AJ, Mahmood F. Fast and scalable search of whole-slide images via self-supervised deep learning. Nat Biomed Eng. 2022;6(12):1420\u201334. https:\/\/doi.org\/10.1038\/s41551-022-00929-8.","journal-title":"Nat Biomed Eng"},{"issue":"1","key":"1710_CR26","doi-asserted-by":"publisher","first-page":"3217","DOI":"10.1038\/s41598-020-59985-2","volume":"10","author":"JD Ianni","year":"2020","unstructured":"Ianni JD, Soans RE, Sankarapandian S, Chamarthi RV, Ayyagari D, Olsen TG, Lee JB. Tailored for real-world: A whole slide image classification system validated on uncurated multi-site data emulating the prospective pathology workload. Sci Rep. 2020;10(1):3217. https:\/\/doi.org\/10.1038\/s41598-020-59985-2.","journal-title":"Sci Rep"},{"key":"1710_CR27","doi-asserted-by":"publisher","unstructured":"Basal cell carcinoma classification [Internet]. AIDA dataset register. Available from: https:\/\/doi.org\/10.23698\/aida\/bccc","DOI":"10.23698\/aida\/bccc"},{"key":"1710_CR28","unstructured":"Geijs D. Classification Of Basal cell carcinoma, Risky skin tumors and Abnormalities [Internet]. Available from: https:\/\/registry.opendata.aws\/cobra\/"},{"key":"1710_CR29","doi-asserted-by":"publisher","first-page":"103063","DOI":"10.1016\/j.media.2023.103063","volume":"93","author":"DJ Geijs","year":"2024","unstructured":"Geijs DJ, Dooper S, Aswolinskiy W, Hillen LM, Amir AL, Litjens G. Detection and subtyping of basal cell carcinoma in whole-slide histopathology using weakly-supervised learning. Med Image Anal. 2024;93:103063.","journal-title":"Med Image Anal"},{"key":"1710_CR30","unstructured":"Chen T, Kornblith S, Norouzi M, Hinton G. A simple framework for contrastive learning of visual representations. In: Proceedings of the 37th International Conference on Machine Learning; 2020 Jul 13\u201318; Virtual. PMLR; 2020. pp. 1597\u20131607."},{"key":"1710_CR31","doi-asserted-by":"publisher","first-page":"281","DOI":"10.1186\/s12859-017-1685-x","volume":"18","author":"Y Xu","year":"2017","unstructured":"Xu Y, Jia Z, Wang LB, Ai Y, Zhang F, Lai M, et al. Large scale tissue histopathology image classification, segmentation, and visualization via deep convolutional activation features. BMC Bioinformatics. 2017;18:281.","journal-title":"BMC Bioinformatics"}],"container-title":["BMC Medical Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12880-025-01710-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s12880-025-01710-4\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s12880-025-01710-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T14:49:27Z","timestamp":1751381367000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmcmedimaging.biomedcentral.com\/articles\/10.1186\/s12880-025-01710-4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,16]]},"references-count":31,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["1710"],"URL":"https:\/\/doi.org\/10.1186\/s12880-025-01710-4","relation":{},"ISSN":["1471-2342"],"issn-type":[{"type":"electronic","value":"1471-2342"}],"subject":[],"published":{"date-parts":[[2025,5,16]]},"assertion":[{"value":"6 September 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"5 May 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 May 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"6 July 2025","order":4,"name":"change_date","label":"Change Date","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"Update","order":5,"name":"change_type","label":"Change Type","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"This article has been updated to amend the license information","order":6,"name":"change_details","label":"Change Details","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The study was conducted in compliance with the Helsinki Declaration. The study was approved by the Swedish Ethical Review Authority (Dnr 2023-03774-01). Informed consent was waived by the Swedish Ethical Review Authority.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not relevant.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"166"}}