{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,25]],"date-time":"2026-08-25T08:28:00Z","timestamp":1787646480806,"version":"build-2736575974"},"reference-count":31,"publisher":"MDPI AG","issue":"17","license":[{"start":{"date-parts":[[2026,8,25]],"date-time":"2026-08-25T00:00:00Z","timestamp":1787616000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Cancers"],"abstract":"<jats:p>Background: Convolutional neural networks (CNNs) perform well in histopathological classification of oral squamous cell carcinoma (OSCC), but their robustness across distinct tissue domains remains insufficiently studied. This study assessed whether a CNN trained only on primary oral and oropharyngeal squamous cell carcinoma images remained transferable to an independent metastatic lymph node histopathology dataset. Methods: Three public datasets containing 14,760 OSCC\/OPSCC and 8530 normal oral mucosa images were combined for model development. An ImageNet-pretrained EfficientNetB0 backbone was used as a fixed feature extractor with a task-specific binary classification head. Performance was first assessed on an independent internal testing subset and subsequently evaluated on 20,000 H&amp;E-stained normal and metastatic lymph node patches from a separate public dataset. Results: The model achieved an internal testing accuracy of 91.48%, with 92.95% sensitivity, 88.92% specificity, 93.56% precision, a 93.25% F1-score, and a Youden\u2019s J index of 0.819. External validation resulted in a substantial decrease in overall classification performance, with an accuracy of 56.91%, specificity of 33.32%, precision of 48.71%, F1-score of 63.37%, balanced accuracy of 61.99%, MCC of 0.279, and a Youden\u2019s J index of 0.240. Nevertheless, sensitivity for metastatic tissue remained high at 90.66%, indicating a markedly asymmetric external error profile characterized predominantly by false-positive classifications. Conclusions: The marked performance decrease during external validation demonstrates the limitations of direct cross-domain transfer between substantially different histopathological environments. However, the preserved sensitivity suggests that some discriminative information remained transferable beyond the development domain. These findings support partial rather than universal cross-domain generalization and emphasize the importance of independent out-of-distribution evaluation when assessing deep learning robustness in computational pathology.<\/jats:p>","DOI":"10.3390\/cancers18172752","type":"journal-article","created":{"date-parts":[[2026,8,25]],"date-time":"2026-08-25T08:00:48Z","timestamp":1787644848000},"page":"2752","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Cross-Domain Generalization of a CNN Trained on Oral and Oropharyngeal Squamous Cell Carcinoma Histopathology Using External Validation on Metastatic Lymph Node Tissue"],"prefix":"10.3390","volume":"18","author":[{"given":"Lorena Adriana","family":"Paun","sequence":"first","affiliation":[{"name":"ENT Department, Carol Davila University of Medicine and Pharmacy, 020021 Bucharest, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Iulian Alexandru","family":"Taciuc","sequence":"additional","affiliation":[{"name":"Pathology Department, Carol Davila University of Medicine and Pharmacy, 020021 Bucharest, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1373-2510","authenticated-orcid":false,"given":"Mihai","family":"Dumitru","sequence":"additional","affiliation":[{"name":"ENT Department, Carol Davila University of Medicine and Pharmacy, 020021 Bucharest, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2347-5661","authenticated-orcid":false,"given":"Daniela","family":"Vrinceanu","sequence":"additional","affiliation":[{"name":"ENT Department, Carol Davila University of Medicine and Pharmacy, 020021 Bucharest, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andreea","family":"Marinescu","sequence":"additional","affiliation":[{"name":"Radiology and Imaging Department, Carol Davila University of Medicine and Pharmacy, 020021 Bucharest, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alexandru-Darius","family":"Dragomir-Serboiu","sequence":"additional","affiliation":[{"name":"ENT Department, Carol Davila University of Medicine and Pharmacy, 020021 Bucharest, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6341-8784","authenticated-orcid":false,"given":"Alina","family":"Oancea","sequence":"additional","affiliation":[{"name":"ENT Department, Elias Clinical Hospital, 011461 Bucharest, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Monica-Mihaela","family":"Cirstoiu","sequence":"additional","affiliation":[{"name":"Gynecology Department, Carol Davila University of Medicine and Pharmacy, 020021 Bucharest, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Adrian","family":"Costache","sequence":"additional","affiliation":[{"name":"Pathology Department, Carol Davila University of Medicine and Pharmacy, 020021 Bucharest, Romania"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,25]]},"reference":[{"key":"ref_1","unstructured":"Winters, R., and Anjum, F. (2026, July 01). Oropharyngeal Squamous Cell Carcinoma, StatPearls, Available online: https:\/\/www.ncbi.nlm.nih.gov\/books\/NBK563268\/."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"S104","DOI":"10.1007\/s12105-012-0368-1","article-title":"Human papillomavirus in non-oropharyngeal head and neck cancers: A systematic literature review","volume":"6","author":"Isayeva","year":"2012","journal-title":"Head Neck Pathol."},{"key":"ref_3","unstructured":"Ferlay, J., Ervik, M., Lam, F., Colombet, M., Mery, L., and Pi\u00f1eros, M. (2018). Global Cancer Observatory: Cancer Today, International Agency for Research on Cancer. Available online: https:\/\/gco.iarc.who.int\/en."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Mart\u00ednez-Barajas, M.G., Jave-Su\u00e1rez, L.F., Ram\u00edrez-L\u00f3pez, I.G., Garc\u00eda-Chagoll\u00e1n, M., Zepeda-Nu\u00f1o, J.S., Ram\u00edrez-de-Arellano, A., Ortiz-Lazareno, P.C., Villegas-Pineda, J.C., and Pereira-Su\u00e1rez, A.L. (2023). HPV-Negative and HPV-Positive Oral Cancer Cells Stimulate the Polarization of Neutrophils towards Different Functional Phenotypes In Vitro. Cancers, 15.","DOI":"10.3390\/cancers15245814"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"982","DOI":"10.1007\/s10147-023-02327-9","article-title":"Association between human papillomavirus and oral cancer: A literature review","volume":"28","author":"Shigeishi","year":"2023","journal-title":"Int. J. Clin. Oncol."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"306","DOI":"10.1038\/s41571-022-00603-7","article-title":"HPV-associated oropharyngeal cancer: Epidemiology, molecular biology and clinical management","volume":"19","author":"Lechner","year":"2022","journal-title":"Nat. Rev. Clin. Oncol."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"909","DOI":"10.1007\/s13193-024-02011-y","article-title":"Histopathological Variants of Oral Squamous Cell Carcinoma Operated at a Cancer Institute in North India","volume":"15","author":"Dey","year":"2024","journal-title":"Indian J. Surg. Oncol."},{"key":"ref_8","first-page":"411","article-title":"Epidemiological and Histopathological Features of Oral Squamous Cell Carcinoma-A Retrospective Study","volume":"50","author":"Ilie","year":"2024","journal-title":"Curr. Health Sci. J."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Dolens, E.D.S., Dourado, M.R., Almangush, A., Salo, T.A., Gurgel Rocha, C.A., da Silva, S.D., Brennan, P.A., and Coletta, R.D. (2021). The Impact of Histopathological Features on the Prognosis of Oral Squamous Cell Carcinoma: A Comprehensive Review and Meta-Analysis. Front. Oncol., 11.","DOI":"10.3389\/fonc.2021.784924"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"107247","DOI":"10.1016\/j.oraloncology.2025.107247","article-title":"Histopathological invasion patterns and prognosis in Oral Squamous Cell Carcinoma: A retrospective analysis of 560 cases","volume":"163","author":"Alfertshofer","year":"2025","journal-title":"Oral Oncol."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"349","DOI":"10.4103\/0973-029X.151311","article-title":"Immunomorphological patterns of cervical lymph nodes in oral squamous cell carcinoma","volume":"18","author":"Chandavarkar","year":"2014","journal-title":"J. Oral Maxillofac. Pathol."},{"key":"ref_12","first-page":"e2021034","article-title":"Well-Differentiated Squamous Cell Carcinoma: Is Histological Differentiation a Relevant Prognostic Parameter?","volume":"11","author":"Boff","year":"2021","journal-title":"Dermatol. Pract. Concept."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"121","DOI":"10.1186\/s13000-023-01407-8","article-title":"Explainable convolutional neural networks for assessing head and neck cancer histopathology","volume":"18","author":"Hecht","year":"2023","journal-title":"Diagn. Pathol."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"1321","DOI":"10.1177\/00220345221089858","article-title":"Histopathology-Based Diagnosis of Oral Squamous Cell Carcinoma Using Deep Learning","volume":"101","author":"Yang","year":"2022","journal-title":"J. Dent. Res."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Bellahsen-Harrar, Y., Lubrano, M., L\u00e9pine, C., Beaufr\u00e8re, A., Bocciarelli, C., Brunet, A., Decroix, E., El-Sissy, F.N., Fabiani, B., and Morini, A. (2025). Exploring the risks of over-reliance on AI in diagnostic pathology. What lessons can be learned to support the training of young pathologists?. PLoS ONE, 20.","DOI":"10.1371\/journal.pone.0323270"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"105114","DOI":"10.1016\/j.dib.2020.105114","article-title":"Histopathological imaging database for oral cancer analysis","volume":"29","author":"Rahman","year":"2020","journal-title":"Data Brief"},{"key":"ref_17","unstructured":"(2026, July 01). Available online: https:\/\/www.kaggle.com\/datasets\/ashenafifasilkebede\/dataset."},{"key":"ref_18","unstructured":"Divine Sebukpor (2026, July 01). Cancer_Classification [Dataset]. Kaggle. Available online: https:\/\/www.kaggle.com\/datasets\/maestroalert\/cancer."},{"key":"ref_19","unstructured":"(2026, July 01). Available online: https:\/\/www.kaggle.com\/datasets\/mohitkumarhalder\/oralnew."},{"key":"ref_20","unstructured":"(2026, July 01). Available online: https:\/\/www.kaggle.com\/datasets\/pankaj705\/histopathologic-cancer-detection."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"giy065","DOI":"10.1093\/gigascience\/giy065","article-title":"1399 H&E-stained sentinel lymph node sections of breast cancer patients: The CAMELYON dataset","volume":"7","author":"Litjens","year":"2018","journal-title":"Gigascience"},{"key":"ref_22","unstructured":"(2026, July 01). Available online: https:\/\/keras.io\/api\/applications\/efficientnet\/efficientnet_models\/#efficientnetb0-function."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"1","DOI":"10.4132\/jptm.2018.12.16","article-title":"Artificial Intelligence in Pathology","volume":"53","author":"Chang","year":"2019","journal-title":"J. Pathol. Transl. Med."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"11485","DOI":"10.1038\/s41467-025-66300-y","article-title":"Knowledge-guided adaptation of pathology foundation models effectively improves cross-domain generalization and demographic fairness","volume":"16","author":"Huang","year":"2025","journal-title":"Nat. Commun."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"6","DOI":"10.4103\/jomfp.jomfp_40_23","article-title":"Lymph nodes in health and disease\u2014A pathologist\u2019s perspective","volume":"27","author":"Priya","year":"2023","journal-title":"J. Oral Maxillofac. Pathol."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"100127","DOI":"10.1016\/j.jpi.2022.100127","article-title":"Impact of scanner variability on lymph node segmentation in computational pathology","volume":"13","author":"Khan","year":"2022","journal-title":"J. Pathol. Inform."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"103155","DOI":"10.1016\/j.media.2024.103155","article-title":"Domain generalization across tumor types, laboratories, and species\u2014Insights from the 2022 edition of the Mitosis Domain Generalization Challenge","volume":"94","author":"Aubreville","year":"2024","journal-title":"Med. Image Anal."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"108018","DOI":"10.1016\/j.compbiomed.2024.108018","article-title":"Automatic data augmentation to improve generalization of deep learning in H&E stained histopathology","volume":"170","author":"Faryna","year":"2024","journal-title":"Comput. Biol. Med."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"LB174","DOI":"10.1158\/1538-7445.AM2026-LB174","article-title":"Abstract LB174: H-optimus-1: A foundation model for computational histopathology","volume":"86","author":"Scalbert","year":"2026","journal-title":"Cancer Res."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Kapse, S., Ayg\u00fcn, M., Cole, E., Lundberg, E., Song, L., and Xing, E.P. (2026). GenBio-PathFM: A State-of-the-Art Foundation Model for Histopathology. bioRxiv.","DOI":"10.64898\/2026.03.17.712534"},{"key":"ref_31","first-page":"640","article-title":"Nasopharyngeal carcinoma\u2014Analysis of risk factors and immunological markers","volume":"107","author":"Anghel","year":"2012","journal-title":"Chirurgia"}],"container-title":["Cancers"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-6694\/18\/17\/2752\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,25]],"date-time":"2026-08-25T08:07:25Z","timestamp":1787645245000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-6694\/18\/17\/2752"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,25]]},"references-count":31,"journal-issue":{"issue":"17","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["cancers18172752"],"URL":"https:\/\/doi.org\/10.3390\/cancers18172752","relation":{},"ISSN":["2072-6694"],"issn-type":[{"value":"2072-6694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,25]]}}}