{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T17:53:27Z","timestamp":1791395607125,"version":"4.3.3"},"reference-count":23,"publisher":"Springer Science and Business Media LLC","license":[{"start":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T00:00:00Z","timestamp":1791331200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T00:00:00Z","timestamp":1791331200000},"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":["J Digit Imaging. Inform. med."],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Pathology foundation models (PFMs) are widely used as frozen encoders, but their behavior under changes in histologic magnification remains poorly characterized. We developed PathScaleBench to quantify cross-scale representation behavior and determine whether it is associated with downstream magnification-shift robustness. Seventy-one audited native-40\u2009\u00d7\u2009The Cancer Genome Atlas whole-slide images from eight cancer types were sampled at the same tissue centers at 40\u2009\u00d7\u2009, 10\u2009\u00d7\u2009, and 2.5\u2009\u00d7\u2009using a native-field-of-view (native-FOV) protocol. Eight PFMs were evaluated using centered kernel alignment (CKA), relative retention, bidirectional cross-scale retrieval, cancer retrieval, and neighborhood preservation. Eight-class linear probes were trained at one magnification and tested at all three using ten repetitions of five-fold patient-grouped cross-validation. Model rankings varied by metric. Phikon achieved the highest mean CKA (0.714), relative retention (0.774), and neighborhood preservation (0.248); Virchow2 achieved the highest cross-scale retrieval similarity (0.643); and Midnight achieved the highest cancer Recall@1 (0.845). No representation metric was significantly associated with Recall@1 at the model level after multiplicity correction. Midnight showed the lowest magnification generalization gap (0.082), followed by Virchow2 (0.181) and UNI2 (0.224). Across 1704 case-model-scale-pair observations, pair-matched retrieval similarity was associated with corresponding shift robustness (Spearman\n                    <jats:italic>\u03c1<\/jats:italic>\n                    \u2009=\u20090.159, 95% confidence interval 0.059\u20130.264; Holm-adjusted\n                    <jats:italic>p<\/jats:italic>\n                    \u2009=\u20090.0021). Cross-scale PFM behavior is multidimensional; global alignment, local correspondence, biological identity, neighborhood topology, and downstream robustness are not interchangeable. No model dominated all dimensions. Representation profiles can support model selection, but magnification robustness should be tested directly.\n                  <\/jats:p>","DOI":"10.1007\/s10278-026-02354-8","type":"journal-article","created":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T17:10:14Z","timestamp":1791393014000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["PathScaleBench: A Multidimensional Benchmark of Cross-Scale Representation Behavior and Magnification-Shift Robustness in Pathology Foundation Models"],"prefix":"10.1007","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6850-9773","authenticated-orcid":false,"given":"Fei","family":"Su","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,10,7]]},"reference":[{"key":"2354_CR1","doi-asserted-by":"publisher","first-page":"775","DOI":"10.1038\/s41591-021-01343-4","volume":"27","author":"J van der Laak","year":"2021","unstructured":"van der Laak J, Litjens G, Ciompi F. Deep learning in histopathology: the path to the clinic. Nat Med. 2021;27:775-784. https:\/\/doi.org\/10.1038\/s41591-021-01343-4","journal-title":"Nat Med."},{"key":"2354_CR2","doi-asserted-by":"publisher","first-page":"930","DOI":"10.1038\/s44222-023-00096-8","volume":"1","author":"AH Song","year":"2023","unstructured":"Song AH, Jaume G, Williamson DFK, et al. Artificial intelligence for digital and computational pathology. Nat Rev Bioeng. 2023;1:930-949. https:\/\/doi.org\/10.1038\/s44222-023-00096-8","journal-title":"Nat Rev Bioeng."},{"key":"2354_CR3","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, et al. Clinical-grade computational pathology using weakly supervised deep learning on whole-slide images. Nat Med. 2019;25:1301-1309. https:\/\/doi.org\/10.1038\/s41591-019-0508-1","journal-title":"Nat Med."},{"key":"2354_CR4","doi-asserted-by":"publisher","DOI":"10.3389\/fpubh.2022.892658","volume":"10","author":"M D'Amato","year":"2022","unstructured":"D'Amato M, Szostak P, Torben-Nielsen B. A comparison between single- and multi-scale approaches for classification of histopathology images. Front Public Health. 2022;10:892658. https:\/\/doi.org\/10.3389\/fpubh.2022.892658","journal-title":"Front Public Health."},{"key":"2354_CR5","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2024.103124","volume":"94","author":"R Deng","year":"2024","unstructured":"Deng R, Cui C, Remedios LW, et al. Cross-scale multi-instance learning for pathological image diagnosis. Med Image Anal. 2024;94:103124. https:\/\/doi.org\/10.1016\/j.media.2024.103124","journal-title":"Med Image Anal."},{"key":"2354_CR6","doi-asserted-by":"crossref","unstructured":"Chen RJ, Chen C, Li Y, et al. Scaling vision transformers to gigapixel images via hierarchical self-supervised learning. In: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition. 2022. p. 16144\u201316155.","DOI":"10.1109\/CVPR52688.2022.01567"},{"key":"2354_CR7","doi-asserted-by":"publisher","first-page":"850","DOI":"10.1038\/s41591-024-02857-3","volume":"30","author":"RJ Chen","year":"2024","unstructured":"Chen RJ, Ding T, Lu MY, et al. Towards a general-purpose foundation model for computational pathology. Nat Med. 2024;30:850-862. https:\/\/doi.org\/10.1038\/s41591-024-02857-3","journal-title":"Nat Med."},{"key":"2354_CR8","doi-asserted-by":"publisher","unstructured":"Filiot A, Ghermi R, Olivier A, et al. Scaling self-supervised learning for histopathology with masked image modeling. medRxiv. 2023. https:\/\/doi.org\/10.1101\/2023.07.21.23292757","DOI":"10.1101\/2023.07.21.23292757"},{"key":"2354_CR9","doi-asserted-by":"publisher","first-page":"2924","DOI":"10.1038\/s41591-024-03141-0","volume":"30","author":"E Vorontsov","year":"2024","unstructured":"Vorontsov E, Bozkurt A, Casson A, et al. A foundation model for clinical-grade computational pathology and rare cancers detection. Nat Med. 2024;30:2924-2935. https:\/\/doi.org\/10.1038\/s41591-024-03141-0","journal-title":"Nat Med."},{"key":"2354_CR10","doi-asserted-by":"publisher","unstructured":"Zimmermann E, Vorontsov E, Viret J, et al. Virchow2: scaling self-supervised mixed magnification models in pathology. arXiv:2408.00738. 2024. https:\/\/doi.org\/10.48550\/arXiv.2408.00738","DOI":"10.48550\/arXiv.2408.00738"},{"key":"2354_CR11","unstructured":"Mahmood Lab. UNI2-h model card. Hugging Face. https:\/\/huggingface.co\/MahmoodLab\/UNI2-h. Accessed 3 September 2026."},{"key":"2354_CR12","doi-asserted-by":"publisher","unstructured":"Karasikov M, van Doorn J, K\u00e4nzig N, et al. Training state-of-the-art pathology foundation models with orders of magnitude less data. In: Medical Image Computing and Computer Assisted Intervention - MICCAI 2025. LNCS 15967. Cham: Springer; 2025. p. 573\u2013583. https:\/\/doi.org\/10.1007\/978-3-032-04984-1_55","DOI":"10.1007\/978-3-032-04984-1_55"},{"key":"2354_CR13","doi-asserted-by":"publisher","first-page":"181","DOI":"10.1038\/s41586-024-07441-w","volume":"630","author":"H Xu","year":"2024","unstructured":"Xu H, Usuyama N, Bagga J, et al. A whole-slide foundation model for digital pathology from real-world data. Nature. 2024;630:181-188. https:\/\/doi.org\/10.1038\/s41586-024-07441-w","journal-title":"Nature."},{"key":"2354_CR14","doi-asserted-by":"publisher","unstructured":"Usuyama N, Valanarasu JMJ, Yao S, et al. GigaPath-Flash and GigaTIME-Flash: efficient pathology foundation models for whole-slide and tumor microenvironment analysis. arXiv:2607.18218. 2026. https:\/\/doi.org\/10.48550\/arXiv.2607.18218","DOI":"10.48550\/arXiv.2607.18218"},{"key":"2354_CR15","doi-asserted-by":"publisher","first-page":"3640","DOI":"10.1038\/s41467-025-58796-1","volume":"16","author":"G Campanella","year":"2025","unstructured":"Campanella G, Chen S, Singh M, et al. A clinical benchmark of public self-supervised pathology foundation models. Nat Commun. 2025;16:3640. https:\/\/doi.org\/10.1038\/s41467-025-58796-1","journal-title":"Nat Commun."},{"key":"2354_CR16","doi-asserted-by":"publisher","first-page":"1113","DOI":"10.1038\/s41551-025-01516-3","volume":"10","author":"P Neidlinger","year":"2026","unstructured":"Neidlinger P, El Nahhas OSM, Muti HS, et al. Benchmarking foundation models as feature extractors for weakly supervised computational pathology. Nat Biomed Eng. 2026;10:1113-1123. https:\/\/doi.org\/10.1038\/s41551-025-01516-3","journal-title":"Nat Biomed Eng."},{"key":"2354_CR17","doi-asserted-by":"publisher","first-page":"5218","DOI":"10.1038\/s41467-026-73923-2","volume":"17","author":"J K\u00f6men","year":"2026","unstructured":"K\u00f6men J, de Jong ED, Hense J, et al. Towards robust foundation models for digital pathology. Nat Commun. 2026;17:5218. https:\/\/doi.org\/10.1038\/s41467-026-73923-2","journal-title":"Nat Commun."},{"key":"2354_CR18","doi-asserted-by":"publisher","unstructured":"M\u00f6llers A, Hense J, Schulz F, Milbich T, Alber M, Ruff L. Mind the Gap: Continuous Magnification Sampling for Pathology Foundation Models. arXiv:2601.02198. 2026. https:\/\/doi.org\/10.48550\/arXiv.2601.02198","DOI":"10.48550\/arXiv.2601.02198"},{"key":"2354_CR19","doi-asserted-by":"publisher","unstructured":"kaiko.ai, Aben N, de Jong ED, Gatopoulos I, K\u00e4nzig N, Karasikov M, et al. Towards Large-Scale Training of Pathology Foundation Models. arXiv:2404.15217. 2024. https:\/\/doi.org\/10.48550\/arXiv.2404.15217","DOI":"10.48550\/arXiv.2404.15217"},{"key":"2354_CR20","first-page":"3519","volume":"97","author":"S Kornblith","year":"2019","unstructured":"Kornblith S, Norouzi M, Lee H, Hinton G. Similarity of neural network representations revisited. Proc Mach Learn Res. 2019;97:3519-3529.","journal-title":"Proc Mach Learn Res."},{"key":"2354_CR21","doi-asserted-by":"publisher","first-page":"1113","DOI":"10.1038\/ng.2764","volume":"45","author":"JN Weinstein","year":"2013","unstructured":"Weinstein JN, Collisson EA, Mills GB, et al. The Cancer Genome Atlas Pan-Cancer analysis project. Nat Genet. 2013;45:1113-1120. https:\/\/doi.org\/10.1038\/ng.2764","journal-title":"Nat Genet."},{"key":"2354_CR22","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102645","volume":"83","author":"X Wang","year":"2023","unstructured":"Wang X, Du Y, Yang S, et al. RetCCL: clustering-guided contrastive learning for whole-slide image retrieval. Med Image Anal. 2023;83:102645. https:\/\/doi.org\/10.1016\/j.media.2022.102645","journal-title":"Med Image Anal."},{"key":"2354_CR23","doi-asserted-by":"publisher","first-page":"3990","DOI":"10.1038\/s41598-025-88545-9","volume":"15","author":"S Alfasly","year":"2025","unstructured":"Alfasly S, Alabtah G, Hemati S, et al. Validation of histopathology foundation models through whole slide image retrieval. Sci Rep. 2025;15:3990. https:\/\/doi.org\/10.1038\/s41598-025-88545-9","journal-title":"Sci Rep."}],"container-title":["Journal of Imaging Informatics in Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10278-026-02354-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10278-026-02354-8","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10278-026-02354-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T17:10:17Z","timestamp":1791393017000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s10278-026-02354-8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,7]]},"references-count":23,"alternative-id":["2354"],"URL":"https:\/\/doi.org\/10.1007\/s10278-026-02354-8","relation":{},"ISSN":["2948-2933"],"issn-type":[{"value":"2948-2933","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,7]]},"assertion":[{"value":"30 July 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"21 September 2026","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"23 September 2026","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"7 October 2026","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The author declares no competing interests.","order":1,"name":"Ethics","label":"Competing interest","group":{"name":"EthicsHeading","label":"Declarations"}}]}}