Self-supervised artificial intelligence system for differentiating mycosis fungoides and benign inflammatory dermatoses
Abstract
Background: Mycosis fungoides (MF), the most common cutaneous T-cell lymphoma, is challenging to diagnose in its early stages because the clinical and histopathological features often overlap with benign inflammatory dermatoses (BIDs), leading to misclassification and delayed treatment. Objectives: To evaluate whether a self-supervised artificial intelligence (AI) system can support diagnostic decision-making in MF and common inflammatory mimics. Methods: This retrospective two-centre study included patients with confirmed diagnoses of MF or BIDs. A self-supervised multimodal system integrating whole-slide histopathology and routine clinical variables was developed to perform multiclass differential classification across MF and common BIDs. Cases were partitioned into training, internal validation and independent external validation cohorts at the patient level. Clinical utility was assessed in a reader study in which dermatopathologists reviewed cases with and without system assistance. Results: Across 786 whole-slide images from 532 patients, the multimodal model achieved macro-area under the curves > 0.85 in both validation sets, with a macro-balanced accuracy of 0.837 [95% confidence interval (CI) 0.778–0.897; n = 106] internally and 0.762 (95% CI 0.724–0.802; n = 260) externally, representing improved performance over the unimodal histopathology model. In the reader study, AI assistance was associated with improved macro-balanced accuracy among junior (from 0.778 to 0.818) and senior (from 0.805 to 0.859) dermatopathologists, accompanied by consistent improvements in macro-averaged sensitivity and specificity across all diagnostic categories. Interpretability analyses generated heatmaps that were generally consistent with recognized histopathological features of MF and BIDs, aligning with dermatopathologist interpretation. Conclusions: This multimodal, self-supervised system may support dermatopathologists by providing interpretable, probability-based guidance for the classification of MF and its common inflammatory mimics.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Zhao et al. demonstrated the feasibility of differentiating MF from BIDs using a self-supervised multimodal framework that integrates H&E whole-slide images with structured clinical variables, achieving a macro-balanced accuracy of 0.84.
“Recent work by Zhao et al. demonstrated the feasibility of differentiating MF from BIDs using a self-supervised multimodal framework that integrates H&E whole-slide images with structured clinical variables, achieving a macro-balanced accuracy of 0·84 18 .”
Zhao et al.'s multiclass algorithm was restricted to differentiating MF from only three specific inflammatory conditions (atopic dermatitis, psoriasis, and drug reactions).
“Furthermore, their multiclass algorithm was restricted to differentiating MF from only three specific inflammatory conditions (atopic dermatitis, psoriasis, and drug reactions).”
Cites
Dermatologist-level classification of skin cancer with deep neural networks
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017) · Nature · doi:10.1038/nature21056 · Cited by 14,362 · see citing works
Masked Autoencoders Are Scalable Vision Learners
He, K., Chen, X., Xie, S., Li, Y., Dollar, P., & Girshick, R. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01553 · Cited by 8,489 · see citing works
WHO-EORTC classification for cutaneous lymphomas
Willemze, R. (2005) · Blood · doi:10.1182/blood-2004-09-3502 · Cited by 3,803 · see citing works
Pathogenesis and clinical features of psoriasis
Griffiths, C. E. M., & Barker, J. N. (2007) · The Lancet · doi:10.1016/s0140-6736(07)61128-3 · Cited by 2,159 · see citing works
Towards a general-purpose foundation model for computational pathology
Chen, R. J., Ding, T., Lu, M. Y., Williamson, D. F. K., Jaume, G., Song, A. H., Chen, B., Zhang, A., Shao, D., Shaban, M., et al. (2024) · Nature Medicine · doi:10.1038/s41591-024-02857-3 · Cited by 1,560 · see citing works
The 2018 update of the WHO-EORTC classification for primary cutaneous lymphomas
Willemze, R., Cerroni, L., Kempf, W., Berti, E., Facchetti, F., Swerdlow, S. H., & Jaffe, E. S. (2019) · Blood · doi:10.1182/blood-2018-11-881268 · Cited by 1,411 · see citing works
WILLIAMS, H. C., JBURNEY, P. G., HAY, R. J., ARCHER, C. B., SHIPLEY, M. J., AHUNTER, J. J., BINGHAM, E. A., FINLAY, A. Y., PEMBROKE, A. C., CGRAHAM-BROWN, R. A., et al. (1994) · British Journal of Dermatology · doi:10.1111/j.1365-2133.1994.tb08530.x · Cited by 1,271 · see citing works
Joint AAD-NPF guidelines of care for the management and treatment of psoriasis with biologics
Menter, A., Strober, B. E., Kaplan, D. H., Kivelevitch, D. N., Prater, E. F., Stoff, B. K., Armstrong, A. W., Connor, C. J., Cordoro, K. M., Davis, D. M. R., et al. (2019) · Journal of the American Academy of Dermatology · doi:10.1016/j.jaad.2018.11.057 · Cited by 972 · see citing works
A whole-slide foundation model for digital pathology from real-world data
Xu, H., Usuyama, N., Bagga, J., Zhang, S., Rao, R., Naumann, T., Wong, C., Gero, Z., González, J., Gu, Y., et al. (2024) · Nature · doi:10.1038/s41586-024-07441-w · Cited by 921 · see citing works
Ständer, S. (2021) · New England Journal of Medicine · doi:10.1056/nejmra2023911 · Cited by 776 · see citing works
A foundation model for clinical-grade computational pathology and rare cancers detection
Vorontsov, E., Bozkurt, A., Casson, A., Shaikovski, G., Zelechowski, M., Severson, K., Zimmermann, E., Hall, J., Tenenholtz, N., Fusi, N., et al. (2024) · Nature Medicine · doi:10.1038/s41591-024-03141-0 · Cited by 572 · see citing works
Defining early mycosis fungoides
Pimpinelli, N., Olsen, E. A., Santucci, M., Vonderheid, E., Haeffner, A. C., Stevens, S., Burg, G., Cerroni, L., Dreno, B., Glusac, E., et al. (2005) · Journal of the American Academy of Dermatology · doi:10.1016/j.jaad.2005.08.057 · Cited by 550 · see citing works
Willemze, R., Hodak, E., Zinzani, P. L., Specht, L., & Ladetto, M. (2018) · Annals of Oncology · doi:10.1093/annonc/mdy133 · Cited by 281 · see citing works
Scarisbrick, J. J., Quaglino, P., Prince, H. M., Papadavid, E., Hodak, E., Bagot, M., Servitje, O., Berti, E., Ortiz‐Romero, P., Stadler, R., et al. (2019) · British Journal of Dermatology · doi:10.1111/bjd.17258 · Cited by 244 · see citing works
Mycosis fungoides: A great imitator
Hodak, E., & Amitay-Laish, I. (2019) · Clinics in Dermatology · doi:10.1016/j.clindermatol.2019.01.004 · Cited by 163 · see citing works
de Masson, A., O’Malley, J. T., Elco, C. P., Garcia, S. S., Divito, S. J., Lowry, E. L., Tawa, M. C., Fisher, D. C., Devlin, P. M., Teague, J. E., et al. (2018) · Science Translational Medicine · doi:10.1126/scitranslmed.aar5894 · Cited by 141 · see citing works
Thurber, S. E., Zhang, B. M., Kim, Y. H., Schrijver, I., Zehnder, J. L., & Köhler, S. (2007) · Journal of the American Academy of Dermatology · doi:10.1016/j.jaad.2007.06.004 · Cited by 125 · see citing works
Progress in the Understanding of the Pathology and Pathogenesis of Cutaneous Drug Eruptions
CROWSON, A. N., Brown, T. J., & MAGRO, C. M. J. (2003) · American Journal of Clinical Dermatology · doi:10.2165/00128071-200304060-00005 · Cited by 79 · see citing works
Epidemiology of cutaneous adverse drug reactions
Mockenhaupt, M. (2017) · Allergologie select · doi:10.5414/alx01508e · Cited by 79 · see citing works
Drug-induced cutaneous pathology
Ramdial, P. K., & Naidoo, D. K. (2009) · Journal of Clinical Pathology · doi:10.1136/jcp.2008.058289 · Cited by 76 · see citing works
Song, S. X., Willemze, R., Swerdlow, S. H., Kinney, M. C., & SAID, J. W. (2013) · American Journal of Clinical Pathology · doi:10.1309/ajcpobdp2oqaj5br · Cited by 69 · see citing works
Real-Life Barriers to Diagnosis of Early Mycosis Fungoides: An International Expert Panel Discussion
Hodak, E., Geskin, L., Guenova, E., Ortiz-Romero, P. L., Willemze, R., Zheng, J., Cowan, R., Foss, F., Mangas, C., & Querfeld, C. (2023) · American Journal of Clinical Dermatology · doi:10.1007/s40257-022-00732-w · Cited by 47 · see citing works
Generating dermatopathology reports from gigapixel whole slide images with HistoGPT
Tran, M., Schmidle, P., Guo, R. R., Wagner, S. J., Koch, V., Lupperger, V., Novotny, B., Murphree, D. H., Hardway, H. D., D’Amato, M., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-60014-x · Cited by 36 · see citing works
Self-citationFukushima‐Nomura, A., Kawasaki, H., Yashiro, K., Obata, S., Tanese, K., Ebihara, T., Saeki, H., Etoh, T., Hasegawa, T., Yazaki, J., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-59340-x · Cited by 33 · see citing works
Schachter, O., Tabibian-Keissar, H., Debby, A., et al. (2020) · Journal of the American Academy of Dermatology · doi:10.1016/j.jaad.2020.05.110 · Cited by 22 · see citing works
Liu, Z., Zhang, Y., Wang, K., Xie, F., & Liu, J. (2025) · British Journal of Dermatology · doi:10.1093/bjd/ljaf212 · Cited by 14 · see citing works
Doeleman, T., Brussee, S., Hondelink, L. M., Westerbeek, D. W. F., Sequeira, A. M., Valkema, P. A., Jansen, P. M., He, J., Vermeer, M. H., Quint, K. D., et al. (2025) · Journal of Investigative Dermatology · doi:10.1016/j.jid.2024.07.036 · Cited by 7 · see citing works
Doeleman, T., Beljaards, E. S. M., Ottevanger, R., Jansen, P. M., Vermeer, M. H., Quint, K. D., Willemze, R., & Schrader, A. M. R. (2025) · British Journal of Dermatology · doi:10.1093/bjd/ljaf184 · Cited by 4 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Doeleman, T., Brussee, S., Valkema, P. A., Kempf, W., Vermeer, M. H., Kers, J., Wynaendts, L. C. D., Kerckhoffs, K. G. P., de Jonge, M. M., Nguyen, A. H., et al. (2026) · medRxiv · doi:10.64898/2026.07.27.26359009
- Authors
- Zhao, J., Bai, J., Li, G., Sun, Q., Li, S., Wang, C., Li, H., Xue, R., & Qiao, J.
- Affiliations
- First Affiliated Hospital Zhejiang University (CN)
- Southern Medical University (CN)
- Venue
- British Journal of Dermatology
- Publisher
- Oxford University Press (OUP)
- Published
- 2026-04-09
- Volume
- 195
- Issue
- 1
- Pages
- 136-145
- ISSN
- 0007-0963 (ISSN-L)
- 1365-2133
- OpenAlex
- W7153852720
- Semantic Scholar
- 50fbba853c0dcba3cac9cc4493e80daa52655478
- Cited by
- 2
- References
- 28
- Citations by year
- 2026: 2
- FWCI
- 13.47
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- British Journal of Dermatology (publishedVersion)
- PubMed (publishedVersion)
- Funders
- National Key Research and Development Program of China (2023YFC2508106)
- Topics
- Health Sciences › Medicine › Dermatology › Cutaneous lymphoproliferative disorders research
- Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Keywords
- mycosis fungoides, cutaneous T-cell lymphoma, differential diagnosis, self-supervised learning, multimodal AI, whole slide images, dermatopathology, AI-assisted diagnosis, interpretability
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
