{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T03:54:50Z","timestamp":1781668490424,"version":"3.54.5"},"reference-count":54,"publisher":"MDPI AG","issue":"6","license":[{"start":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T00:00:00Z","timestamp":1781222400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["MAKE"],"abstract":"<jats:p>Foundation models (FMs) are increasingly proposed as general-purpose solutions for computational pathology, with the potential to simplify clinical artificial intelligence deployment by reducing the need for task-specific architectures. However, their reliability across cancer domains with distinct morphological characteristics remains unclear, limiting confidence in real-world clinical use. We benchmarked seven general-purpose pathology FMs and three domain-specific FMs across eleven patch-level datasets spanning three clinically relevant domains: pediatric hematology, prostate cancer, and breast cancer, using both linear probing and last-layer fine-tuning adaptation strategies. By jointly evaluating pediatric leukemia, male-predominant prostate cancer, and female-predominant breast cancer, this study is, to our knowledge, the first to explicitly examine specialist-versus-generalist FM behavior across age- and sex-stratified cancer populations. Performance differences were strongly domain dependent. In hematology, the specialist FM DINOBloom matched and, in several datasets, marginally exceeded leading generalist models (AUC 0.990\u20130.999 vs. GigaPath 0.981\u20131.000), suggesting advantages for highly distinctive cellular morphology. In prostate cancer grading, the generalist FM UNI2-h consistently outperformed the specialist HistoEncoder (AUC 0.956\u20130.977 vs. 0.908\u20130.964). In breast cancer, UNI2-h achieved the best overall performance across all tasks. No publicly available breast-cancer-specific FM currently exists for direct comparison; therefore, breast cancer results characterize general FM transferability rather than specialist-versus-generalist differences. Importantly, cross-dataset experiments revealed substantial performance degradation under dataset shift in both prostate and breast cancer, indicating that current FMs are not yet robust enough for heterogeneous multi-site clinical use. These findings support the use of generalist FMs as efficient backbones for well-characterized single-site, patch-level tasks, while challenging the assumption that high benchmark performance necessarily reflects true clinical readiness and demonstrating that pathology FMs are not uniformly superior to specialist models.<\/jats:p>","DOI":"10.3390\/make8060164","type":"journal-article","created":{"date-parts":[[2026,6,15]],"date-time":"2026-06-15T00:44:58Z","timestamp":1781484298000},"page":"164","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Do Foundation Models Truly Outperform Domain-Specific Models? Evidence from Digital Pathology"],"prefix":"10.3390","volume":"8","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4153-3287","authenticated-orcid":false,"given":"Chaima","family":"Ben Rabah","sequence":"first","affiliation":[{"name":"AI Innovation Lab, Weill Cornell Medicine, Doha P.O. Box 24144, Qatar"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4145-5509","authenticated-orcid":false,"given":"Ahmed","family":"Serag","sequence":"additional","affiliation":[{"name":"AI Innovation Lab, Weill Cornell Medicine, Doha P.O. Box 24144, Qatar"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,6,12]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"2199","DOI":"10.1001\/jama.2017.14585","article-title":"Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer","volume":"318","author":"Bejnordi","year":"2017","journal-title":"JAMA"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1559","DOI":"10.1038\/s41591-018-0177-5","article-title":"Classification and mutation prediction from non\u2013small cell lung cancer histopathology images using deep learning","volume":"24","author":"Coudray","year":"2018","journal-title":"Nat. Med."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"575","DOI":"10.1038\/s41591-022-01709-2","article-title":"Deep learning-enabled assessment of cardiac allograft rejection from endomyocardial biopsies","volume":"28","author":"Lipkova","year":"2022","journal-title":"Nat. Med."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"7193","DOI":"10.1038\/s41598-018-24876-0","article-title":"A cluster-then-label semi-supervised learning approach for pathology image classification","volume":"8","author":"Peikari","year":"2018","journal-title":"Sci. Rep."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"118790","DOI":"10.1016\/j.neuroimage.2021.118790","article-title":"Deep learning for Alzheimer\u2019s disease: Mapping large-scale histological tau protein for neuroimaging biomarker validation","volume":"248","author":"Ushizima","year":"2022","journal-title":"Neuroimage"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"21","DOI":"10.1186\/s40478-022-01318-7","article-title":"Antemortem detection of Parkinson\u2019s disease pathology in peripheral biopsies using artificial intelligence","volume":"10","author":"Signaevsky","year":"2022","journal-title":"Acta Neuropathol. Commun."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"e253","DOI":"10.1016\/S1470-2045(19)30154-8","article-title":"Digital pathology and artificial intelligence","volume":"20","author":"Niazi","year":"2019","journal-title":"Lancet Oncol."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Serag, A., Ion-Margineanu, A., Qureshi, H., McMillan, R., Saint Martin, M.J., Diamond, J., O\u2019Reilly, P., and Hamilton, P. (2019). Translational AI and deep learning in diagnostic pathology. Front. Med., 6.","DOI":"10.3389\/fmed.2019.00185"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"101544","DOI":"10.1016\/j.media.2019.101544","article-title":"Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology","volume":"58","author":"Tellez","year":"2019","journal-title":"Med. Image Anal."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"325","DOI":"10.1109\/JBHI.2020.3032060","article-title":"Measuring domain shift for deep learning in histopathology","volume":"25","author":"Stacke","year":"2020","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"2307","DOI":"10.1038\/s41591-023-02504-3","article-title":"A visual\u2013language foundation model for pathology image analysis using medical twitter","volume":"29","author":"Huang","year":"2023","journal-title":"Nat. Med."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"259","DOI":"10.1038\/s41586-023-05881-4","article-title":"Foundation models for generalist medical artificial intelligence","volume":"616","author":"Moor","year":"2023","journal-title":"Nature"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"850","DOI":"10.1038\/s41591-024-02857-3","article-title":"Towards a general-purpose foundation model for computational pathology","volume":"30","author":"Chen","year":"2024","journal-title":"Nat. Med."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Tizhoosh, H.R. (2025). Why Foundation Models in Pathology Are Failing. arXiv.","DOI":"10.1038\/s41551-026-01696-6"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"557","DOI":"10.1038\/s41746-025-01926-2","article-title":"Robustness tests for biomedical foundation models should tailor to specifications","volume":"8","author":"Xian","year":"2025","journal-title":"npj Digit. Med."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Rabah, C.B., and Serag, A. (2025). Beyond Broad Applications: Can Pathology Foundation Models Adapt to Hematopathology?. Proceedings of the International Workshop on Foundation Models for General Medical AI, Springer.","DOI":"10.1007\/978-3-032-07845-2_13"},{"key":"ref_17","unstructured":"Marza, P., Fillioux, L., Boutaj, S., Mahatha, K., Desrosiers, C., Piantanida, P., Dolz, J., Christodoulidis, S., and Vakalopoulou, M. (2025). THUNDER: Tile-level Histopathology image UNDERstanding benchmark. arXiv."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Bareja, R., Carrillo-Perez, F., Zheng, Y., Pizurica, M., Nandi, T.N., Shen, J., Madduri, R., and Gevaert, O. (medRxiv, 2025). Evaluating Vision and Pathology Foundation Models for Computational Pathology: A Comprehensive Benchmark Study, medRxiv, preprint.","DOI":"10.1101\/2025.05.08.25327250"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Neidlinger, P., El Nahhas, O.S., Muti, H.S., Lenz, T., Hoffmeister, M., Brenner, H., van Treeck, M., Langer, R., Dislich, B., and Behrens, H.M. (2025). Benchmarking foundation models as feature extractors for weakly supervised computational pathology. Nat. Biomed. Eng., 1\u201311.","DOI":"10.1038\/s41551-025-01516-3"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"3640","DOI":"10.1038\/s41467-025-58796-1","article-title":"A clinical benchmark of public self-supervised pathology foundation models","volume":"16","author":"Campanella","year":"2025","journal-title":"Nat. Commun."},{"key":"ref_21","unstructured":"Ma, J., Xu, Y., Zhou, F., Wang, Y., Jin, C., Guo, Z., Wu, J., Tang, O.K., Zhou, H., and Wang, X. (2025). Pathbench: A comprehensive comparison benchmark for pathology foundation models towards precision oncology. arXiv."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Peng, Z., Ayad, M.A., Jing, Y., Chou, T., Cooper, L.A., and Goldstein, J.A. (medRxiv, 2025). Benchmarking pathology foundation models for non-neoplastic pathology in the placenta, medRxiv, preprint.","DOI":"10.1101\/2025.03.19.25324282"},{"key":"ref_23","unstructured":"Aria, M., Ghaderzadeh, M., Bashash, D., Abolghasemi, H., Asadi, F., and Hosseini, A. (2026, June 06). Acute Lymphoblastic Leukemia (ALL) Image Dataset. Available online: https:\/\/www.kaggle.com\/datasets\/mehradaria\/leukemia."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"105474","DOI":"10.1016\/j.dib.2020.105474","article-title":"A dataset for microscopic peripheral blood cell images for development of automatic recognition systems","volume":"30","author":"Acevedo","year":"2020","journal-title":"Data Brief"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"466","DOI":"10.1038\/s41597-023-02378-7","article-title":"A high-resolution large-scale dataset of pathological and normal white blood cells","volume":"10","author":"Bodzas","year":"2023","journal-title":"Sci. Data"},{"key":"ref_26","unstructured":"Matek, C., Schwarz, S., Marr, C., and Spiekermann, K. (2019). A Single-Cell Morphological Dataset of Leukocytes from AML Patients and Non-Malignant Controls [Data Set], The Cancer Imaging Archive (TCIA)."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"105714","DOI":"10.1016\/j.compbiomed.2022.105714","article-title":"Proportion constrained weakly supervised histopathology image classification","volume":"147","author":"Schmidt","year":"2022","journal-title":"Comput. Biol. Med."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"108472","DOI":"10.1016\/j.cmpb.2024.108472","article-title":"The CrowdGleason dataset: Learning the Gleason grade from crowds and experts","volume":"257","author":"Morquecho","year":"2024","journal-title":"Comput. Methods Programs Biomed."},{"key":"ref_29","first-page":"2020","article-title":"SICAPv2-prostate whole slide images with gleason grades annotations","volume":"1","year":"2020","journal-title":"Mendeley Data"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"1455","DOI":"10.1109\/TBME.2015.2496264","article-title":"A dataset for breast cancer histopathological image classification","volume":"63","author":"Spanhol","year":"2015","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"ref_31","unstructured":"Mooney, P. (2025, November 30). Breast Histopathology Images. Available online: https:\/\/www.kaggle.com\/datasets\/paultimothymooney\/breast-histopathology-images."},{"key":"ref_32","unstructured":"Valieris, R., Martins, L., Defelicibus, A., de Toledo Osorio, C.A.B., and Bueno, A.P. (2026, June 06). 2 Million Histological Images of Breast Cancer Tumors with HER2 Labels. Available online: https:\/\/zenodo.org\/records\/8383580."},{"key":"ref_33","unstructured":"Nechaev, D., Pchelnikov, A., and Ivanova, E. (2024). Hibou: A family of foundational vision transformers for pathology. arXiv."},{"key":"ref_34","unstructured":"Filiot, A., Jacob, P., Mac Kain, A., and Saillard, C. (2024). Phikon-v2, a large and public feature extractor for biomarker prediction. arXiv."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"181","DOI":"10.1038\/s41586-024-07441-w","article-title":"A whole-slide foundation model for digital pathology from real-world data","volume":"630","author":"Xu","year":"2024","journal-title":"Nature"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"695","DOI":"10.1038\/s41746-025-02027-w","article-title":"Pathorchestra: A comprehensive foundation model for computational pathology with over 100 diverse clinical-grade tasks","volume":"8","author":"Yan","year":"2025","journal-title":"npj Digit. Med."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"863","DOI":"10.1038\/s41591-024-02856-4","article-title":"A visual-language foundation model for computational pathology","volume":"30","author":"Lu","year":"2024","journal-title":"Nat. Med."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Zedda, L., Loddo, A., Di Ruberto, C., and Marr, C. (2025). RedDino: A foundation model for red blood cell analysis. Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer.","DOI":"10.1007\/978-3-032-04965-0_42"},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Koch, V., Wagner, S.J., Kazeminia, S., Sancar, E., Hehr, M., Schnabel, J.A., Peng, T., and Marr, C. (2024). DinoBloom: A foundation model for generalizable cell embeddings in hematology. Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer.","DOI":"10.1007\/978-3-031-72390-2_49"},{"key":"ref_40","unstructured":"Pohjonen, J., Batouche, A.O., Rannikko, A., Sandeman, K., Erickson, A., Pitkanen, E., and Mirtti, T. (2024). HistoEncoder: A digital pathology foundation model for prostate cancer. arXiv."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Koonce, B. (2021). ResNet 50. Convolutional Neural Networks with Swift for Tensorflow: Image Recognition and Dataset Categorization, Springer.","DOI":"10.1007\/978-1-4842-6168-2"},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"87","DOI":"10.1109\/TPAMI.2022.3152247","article-title":"A survey on vision transformer","volume":"45","author":"Han","year":"2022","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., and Guo, B. (2021). Swin transformer: Hierarchical vision transformer using shifted windows. Proceedings of the IEEE\/CVF International Conference on Computer Vision, IEEE.","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Liu, Z., Mao, H., Wu, C.Y., Feichtenhofer, C., Darrell, T., and Xie, S. (2022). A convnet for the 2020s. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR52688.2022.01167"},{"key":"ref_45","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"},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Corley, I., Robinson, C., Dodhia, R., Ferres, J.M.L., and Najafirad, P. (2024). Revisiting pre-trained remote sensing model benchmarks: Resizing and normalization matters. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPRW63382.2024.00322"},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Liang, Y., Zhu, L., Wang, X., and Yang, Y. (2022). A simple episodic linear probe improves visual recognition in the wild. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR52688.2022.00934"},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"151","DOI":"10.1038\/s41698-024-00652-4","article-title":"Learning generalizable AI models for multi-center histopathology image classification","volume":"8","author":"Darbandsari","year":"2024","journal-title":"npj Precis. Oncol."},{"key":"ref_49","unstructured":"Bommasani, R., Hudson, D.A., Adeli, E., Altman, R., Arora, S., von Arx, S., Bernstein, M.S., Bohg, J., Bosselut, A., and Brunskill, E. (2021). On the opportunities and risks of foundation models. arXiv."},{"key":"ref_50","unstructured":"Zimmermann, E., Vorontsov, E., Viret, J., Casson, A., Zelechowski, M., Shaikovski, G., Tenenholtz, N., Hall, J., Klimstra, D., and Yousfi, R. (2024). Virchow2: Scaling self-supervised mixed magnification models in pathology. arXiv."},{"key":"ref_51","unstructured":"Saillard, C., Jenatton, R., Llinares-L\u00f3pez, F., Mariet, Z., Cahan\u00e9, D., Durand, E., and Vert, J.P. (2026, June 06). H-optimus-0 (2024). Available online: https:\/\/github.com\/bioptimus\/releases\/tree\/main\/models\/h-Optimus\/v0."},{"key":"ref_52","unstructured":"Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., and Gelly, S. (2020). An image is worth 16x16 words: Transformers for image recognition at scale. arXiv."},{"key":"ref_53","doi-asserted-by":"crossref","first-page":"109173","DOI":"10.1016\/j.bspc.2025.109173","article-title":"MammXAI: An XAI integrated adaptive multi-model deep learning approach for breast cancer detection using multi-modality images","volume":"113","author":"Singh","year":"2026","journal-title":"Biomed. Signal Process. Control"},{"key":"ref_54","doi-asserted-by":"crossref","unstructured":"Ben Rabah, C., Sattar, A., Ibrahim, A., and Serag, A. (2025). A multimodal deep learning model for the classification of breast cancer subtypes. Diagnostics, 15.","DOI":"10.3390\/diagnostics15080995"}],"container-title":["Machine Learning and Knowledge Extraction"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2504-4990\/8\/6\/164\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T03:09:49Z","timestamp":1781665789000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2504-4990\/8\/6\/164"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,12]]},"references-count":54,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["make8060164"],"URL":"https:\/\/doi.org\/10.3390\/make8060164","relation":{},"ISSN":["2504-4990"],"issn-type":[{"value":"2504-4990","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,12]]}}}