The DOI matches its Crossref record; 186 of the 186 references it cites were found in a registry or an index.
Towards a general-purpose foundation model for computational pathology
Open accessAbstract
Quantitative evaluation of tissue images is crucial for computational pathology (CPath) tasks, requiring the objective characterization of histopathological entities from whole-slide images (WSIs). The high resolution of WSIs and the variability of morphological features present significant challenges, complicating the large-scale annotation of data for high-performance applications. To address this challenge, current efforts have proposed the use of pretrained image encoders through transfer learning from natural image datasets or self-supervised learning on publicly available histopathology datasets, but have not been extensively developed and evaluated across diverse tissue types at scale. We introduce UNI, a general-purpose self-supervised model for pathology, pretrained using more than 100 million images from over 100,000 diagnostic H&E-stained WSIs (>77 TB of data) across 20 major tissue types. The model was evaluated on 34 representative CPath tasks of varying diagnostic difficulty. In addition to outperforming previous state-of-the-art models, we demonstrate new modeling capabilities in CPath such as resolution-agnostic tissue classification, slide classification using few-shot class prototypes, and disease subtyping generalization in classifying up to 108 cancer types in the OncoTree classification system. UNI advances unsupervised representation learning at scale in CPath in terms of both pretraining data and downstream evaluation, enabling data-efficient artificial intelligence models that can generalize and transfer to a wide range of diagnostically challenging tasks and clinical workflows in anatomic pathology. Pretrained using over 100,000 diagnostic histopathological slides across 20 major tissue types, a self-supervised model is shown to outperform existing baselines across various clinically relevant computational pathology tasks.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“By increasing the patch size from the widely used 256 × 256 pixels, we effectively reduce the sequence length by four without impacting the representation quality due to higher resolution patch input, leveraging the robustness of the patch-level foundation models in generalizing to higher resolutions9,10,87.”
From A multimodal whole-slide foundation model for pathology (2025) · cites this paper as [9]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“we present Virchow, the largest foundation model for computational pathology to date.”
From A foundation model for clinical-grade computational pathology and rare cancers detection (2024) · cites this paper as [41]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“To overcome these limitations, several research groups have alternatively proposed the use of weakly-supervised deep learning models, which can be trained to perform various downstream computational pathology tasks such as tumor subtyping and prognosis32–37 without any region or pixel-level annotations.”
From Predicting the tumor microenvironment composition and immunotherapy response in non-small cell lung cancer from digital histopathology images (2024) · cites this paper as [32]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“UNI [191], Prov-GigaPath [192], and Virchow [193] were pretrained solely on pathology image data to perform cancer classification, diagnosis, and prognosis.”
From Challenges in AI-driven Biomedical Multimodal Data Fusion and Analysis (2025) · cites this paper as [191]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Built on rich knowledge representations, these models have shown impressive performance across medical fields such as ophthalmology20, radiology21 and pathology22–25.”
From A Multimodal Vision Foundation Model for Clinical Dermatology (2025) · cites this paper as [23]
The cited source says what the claim attributes to it.
In the citing paper
“The vision module is based on the patch encoder CTransPath17 for the small and medium models, and UNI18 for the large model.”
From Generating dermatopathology reports from gigapixel whole slide images with HistoGPT (2025) · cites this paper as [18]
The cited source says what the claim attributes to it.
In the citing paper
“We found out that the UNI foundation model32, when paired with multi-task AB-MIL, achieved the best predictive performance for the various TME signature prediction tasks.”
From Predicting the tumor microenvironment composition and immunotherapy response in non-small cell lung cancer from digital histopathology images (2024) · cites this paper as [32]
In the citing paper
“In fact GMA is widely used to assess the performance of pathology foundation models23,25,26.”
From A clinical benchmark of public self-supervised pathology foundation models (2025) · cites this paper as [23]
In the citing paper
“Self-supervised learning has become a widely used approach for pretraining foundation models on thousands to tens of thousands of images without annotations or labels, marking an important paradigm shift in the field.20,21”
From Use of artificial intelligence for liver diseases: A survey from the EASL congress 2024 (2024) · cites this paper as [20]
In the citing paper
“An increasing number of published studies show the potential of foundation models20,24; however, they are currently underrepresented in the field of liver diseases.4”
From Use of artificial intelligence for liver diseases: A survey from the EASL congress 2024 (2024) · cites this paper as [20]
In the citing paper
“These works demonstrate that image features produced with self-supervised learning of pathology images outperform image features trained on natural images and that performance improves with scale.”
From A foundation model for clinical-grade computational pathology and rare cancers detection (2024) · cites this paper as [41]
In the citing paper
“Despite the challenges in collecting large-scale datasets in the pathology domain, recent pioneering works have utilized datasets ranging from 30,000 to 400,000 WSIs to train foundation models ranging in size from 28 million to 307 million parameters35–42”
From A foundation model for clinical-grade computational pathology and rare cancers detection (2024) · cites this paper as [41]
In the citing paper
“Specialty-specific foundation models20,21,23 typically focus on single imaging modalities”
From A Multimodal Vision Foundation Model for Clinical Dermatology (2025) · cites this paper as [23]
In the citing paper
“The scaling behavior observed in PanDerm’s performance aligns with recent foundation model trends20,22,23,59”
From A Multimodal Vision Foundation Model for Clinical Dermatology (2025) · cites this paper as [23]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Scikit-learn: Machine Learning in Python
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., et al. (2012) · arXiv · doi:10.48550/arxiv.1201.0490 · Cited by 71,016
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., & Guo, B. (2021) · 2021 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00986 · Cited by 35,534
PyTorch: An Imperative Style, High-Performance Deep Learning Library
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., et al. (2019) · arXiv · doi:10.48550/arxiv.1912.01703 · Cited by 31,473
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Advances in neural information processing systems 7
Cios, K. J., & Shields, M. E. (1997) · Neurocomputing · doi:10.1016/s0925-2312(97)90018-7 · Cited by 22,093
Least squares quantization in PCM
Lloyd, S. (1982) · IEEE Transactions on Information Theory · doi:10.1109/tit.1982.1056489 · Cited by 15,836
Goldberger, A. L., Amaral, L. A. N., Glass, L., Hausdorff, J. M., Ivanov, P. C., Mark, R. G., Mietus, J. E., Moody, G. B., Peng, C.-K., & Stanley, H. E. (2000) · Circulation · doi:10.1161/01.cir.101.23.e215 · Cited by 14,809
Research works printed but not linked: 13
- Chen, T., Kornblith, S., Norouzi, M. & Hinton, G. A simple framework for contrastive learning of visual representations. In International Conference on Machine Learning, 1597–1607 (PMLR, 2020).
- Chen, R. J. & Krishnan, R. G. Self-supervised vision transformers learn visual concepts in histopathology. In Learning Meaningful Representations of Life, NeurIPS 2021 (2022).
- Dosovitskiy, A. et al. An image is worth 16×16 words: transformers for image recognition at scale. In International Conference on Learning Representations (2021).
- Xiang, J. & Zhang, J. Exploring low-rank property in multiple instance learning for whole slide image classification. In The Eleventh International Conference on Learning Representations (2022).
- Wang, X. et al. SCL-WC: cross-slide contrastive learning for weakly-supervised whole-slide image classification. Advances in Neural Information Processing Systems 35, 18009–18021 (2022).
- Zhou, J. et al. iBOT: image BERT pre-training with online tokenizer. In International Conference on Learning Representations (2022).
- Devlin, J., Chang, M.-W., Lee, K. & Toutanova, K. BERT: pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) (2018).
- Sablayrolles, A., Douze, M., Schmid, C. & Jégou, H. Spreading vectors for similarity search. In International Conference on Learning Representations (2019).
- Touvron, H., Vedaldi, A., Douze, M. & Jegou, H. Fixing the train–test resolution discrepancy. In Advances in Neural Information Processing Systems, Vol. 32 (eds Wallach, H. et al.) (Curran Associates, 2019).
- Dao, T., Fu, D. Y., Ermon, S., Rudra, A. & Ré, C. FlashAttention: fast and memory-efficient exact attention with IO-awareness. In Advances in Neural Information Processing Systems (2022).
- Loshchilov, I. & Hutter, F. Decoupled weight decay regularization. In International Conference on Learning Representations (2019).
- Sarıyıldız, M. B., Kalantidis, Y., Alahari, K. & Larlus, D. No reason for no supervision: improved generalization in supervised models. In The Eleventh International Conference on Learning Representations (2023).
- Yang, J., Chen, H., Yan, J., Chen, X. & Yao, J. Towards better understanding and better generalization of low-shot classification in histology images with contrastive learning. In International Conference on Learning Representations (2021).
Cited by
Sorted by citations
Items 401–500 of 1,003
Shi, Z., Xue, S., & Min, W. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.112647 · Cited by 4
Kwon, J., & Shin, Y. (2025) · European Journal of Neuroscience · doi:10.1111/ejn.70376 · Cited by 5
UniCAS: A foundation model for cervical cytology screening
Jiang, H., Cai, J., Shen, Z., Xu, M., Fei, M., Huang, H., Wang, X., Bi, R., Shen, D. G., Zhang, L., et al. (2026) · Cell Reports Medicine · doi:10.1016/j.xcrm.2025.102570 · Cited by 5
Chai, E., Gu, X., Lu, Z., & Cui, T. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.131298 · Cited by 4
Ngassam, B. T., Niu, H., Pang, S., Shydlouskaya, V., & Andrews, T. S. (2026) · Frontiers in Bioinformatics · doi:10.3389/fbinf.2025.1715821 · Cited by 4
Qiqi, K., & Ban, Y. (2026) · Frontiers in Oncology · doi:10.3389/fonc.2026.1745519 · Cited by 7
Hezil, N., Bouridane, A., Hamoudi, R., Al-Maadeed, S. A., Akbari, Y., & Abdullakutty, F. C. (2026) · Medical Engineering & Physics · doi:10.1088/1873-4030/ae449b · Cited by 6
“Recent advances have introduced foundation models pre-trained on millions of histopathology images (Chen et al 2024, Lu et al 2024), demonstrating remarkable performance across diverse pathology tasks.”
Foundation models in medical imaging: A review
Chen, Q., Liu, A., Zhang, J., Yang, C., & Zhang, Y. (2026) · EngMedicine · doi:10.1016/j.engmed.2026.100123 · Cited by 8
Liu, K., Luo, Z., Zhang, W., Pan, Q., Su, X., Yang, Z., Zhang, Q., Wang, B., Tang, B., he, Z., et al. (2026) · Molecular Cancer · doi:10.1186/s12943-026-02589-7 · Cited by 7
“…can be used to verify whether the model focuses on regions consistent with pathologists’ interpretation, identify micrometastatic foci or biomarkers, and further reveal associations between highlighted regions and adverse prognosis or therapeutic response in tumor patients [53, 54, 59, 107].”
Tu, P., Zheng, C., Xie, X., Lv, J., Xie, M., Guo, J., Yin, S., Qiu, K., Wei, Y., Wang, C., et al. (2026) · Nature Biomedical Engineering · doi:10.1038/s41551-026-01622-w · Cited by 4
Wu, S., Hong, G., Wang, Y., Zeng, H., Lin, Z., Yang, J., Chen, J., Chen, H., Huang, Y., Hong, L., et al. (2026) · The Lancet Digital Health · doi:10.1016/j.landig.2025.100961 · Cited by 4
Self-citationA multi-modal foundation model for brain disease diagnosis and medical imaging
Zhang, G., Gao, Z., Duan, C., Liu, J., Lizhu, Y., Liu, Y., Chen, Q., Wang, L., Fei, K., Wang, T., et al. (2026) · Patterns · doi:10.1016/j.patter.2026.101538 · Cited by 4
Priyanka, V., Narendra, M., & Rao, T. D. (2026) · Frontiers in Digital Health · doi:10.3389/fdgth.2026.1752938 · Cited by 4
“Recent work [23] demonstrates that such models can achieve strong performance across diverse histopathological benchmarks.”
Silberman, N., Parwani, A. V., McClintock, D. S., Lujan, G. M., Gondim, D. D., Garcia, C., Hanna, M. G., Pantanowitz, L., Lennerz, J. K., Showalter, T. N., et al. (2026) · AI in Precision Oncology · doi:10.1177/2993091x261455975 · Cited by 5
Health system learning enables generalist neuroimaging models
Kondepudi, A., Rao, A. T., Zhao, C., Lyu, Y., Harake, E. S., Banerjee, S., Ogle, J. D., Joshi, R. S., Meißner, A.-K., Hou, X., et al. (2026) · Nature Medicine · doi:10.1038/s41591-026-04497-1 · Cited by 17
“AB-MIL is known to assign high attention to diagnostic regions in medical images 35 .”
The Virtual Tissues foundation model resolves spatial proteomics across scales
Wenckstern, J., Jain, E., von Querfurth, B., Cheng, Y., Vasilev, K., Pariset, M., Cheng, P. F., Liakopoulos, P., Michielin, O. A., WICKI, A. N., et al. (2026) · Nature · doi:10.1038/s41586-026-10884-y · Cited by 21
Yadav, A., Alvarez, K., Chechenina, A., Yip, K. Y.-L., Ruppin, E., Colas, A. R., Maher, J. Y., Gomez‐Lobo, V., Kumsta, C., & Sinha, S. (2026) · Nature Aging · doi:10.1038/s43587-026-01200-4 · Cited by 6
“24 and 25), PC analysis-based trajectories exhibited smoother and more consistent age-dependent patterns, indicating that low-dimensional representations better capture underlying molecular aging dynamics compared with raw feature-level analyses.”
Phatak, S. B., Saptarshi, R., Sharma, V., Shah, R., Zanwar, A. A., Hegde, P., Chakraborty, S., & Goel, P. (2024) · Lara D. Veeken · doi:10.1093/rheumatology/keae678 · Cited by 4
He, Y., Yang, X., Ye, S., Wen, Z., Liu, P., Li, H., Xiang, F., Liao, J., Tan, Y., Ning, H., et al. (2025) · APL Photonics · doi:10.1063/5.0244272 · Cited by 4
ALPaCA: Adapting Llama for Pathology Context Analysis to enable slide-level question answering
Gao, Z., He, K., Su, W., Pang, X., Machado, I. P., Jimenez‐Liñan, M., Rous, B., Wang, C., Li, C., McGough, W., et al. (2026) · Nature Communications · doi:10.1038/s41467-026-76372-z · Cited by 6
“Speci fi cally, we tested alternative pathology foundation model encoders (UNI 19 , MUSK 43 , PathGen 44 , Supplementary Note 1: Patch encoders) and included text-only and random-image baselines (Supplementary Note 2: Baseline controls) to assess the contribution of visual features.”
Self-citationCellOMaps: A compact representation for robust classification of lung adenocarcinoma growth patterns
Al-Rubaian, A., Gunesli, G. N., Althakfi, W., Azam, A. S., Snead, D. R. J., Rajpoot, N., & Raza, S. E. A. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.110127 · Cited by 4
Deep learning-based interpretable prediction of recurrence of diffuse large B-cell lymphoma
Naji, H., Hahn, P., Pisula, J. I., Ugliano, S., Simon, A., Büttner, R., & Bożek, K. (2025) · BJC Reports · doi:10.1038/s44276-025-00147-0 · Cited by 5
Revisiting Likelihood-Based Out-of-Distribution Detection by Modeling Representations
Ding, Y., Aleksandraus, A., Ahmadian, A., Unger, J., Lindsten, F., & Eilertsen, G. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-95918-9_12 · Cited by 4
Now and future of artificial intelligence-based signet ring cell diagnosis: A survey
Zhu, M. S., Dong, J., Guo, L., Su, F., Liu, J., Wang, G., & Zhao, Z. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.129188 · Cited by 6
Completing spatial transcriptomics data for gene expression prediction benchmarking
Ruiz, D., Cárdenas, P., Manrique, L., Vega, D., Mejia, G., & Arbeláez, P. (2025) · Medical Image Analysis · doi:10.1016/j.media.2025.103754 · Cited by 4
AI-driven WHO 2021 classification of gliomas based only on H&E-stained slides
Innani, S., Bell, W. R., Nasrallah, M. P., Baheti, B., & Bakas, S. (2025) · Neuro-Oncology · doi:10.1093/neuonc/noaf189 · Cited by 6
PathBot: A Foundation Model for Pathological Image Analysis
Lu, M., Wang, T., Zeng, Q., Lu, Z., Li, Z., & Xia, Y. (2025) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2025.3619967 · Cited by 4
“Fourth, evaluation efforts primarily emphasize classification tasks, with limited attention to spatially intensive applications like segmentation and detection [13], [14].”
InfluentialSelf-citationLarge Language Models for Translational Cancer Informatics
Pan, Y., Wang, Y., Wang, G., Su, J., Topaloĝlu, Ü., & Song, Q. (2025) · JCO Clinical Cancer Informatics · doi:10.1200/cci-25-00108 · Cited by 5
Zhou, F., Xu, Y., Cui, Y., Zhang, S., Zhu, Y., He, W., Zhang, D., Wang, J., Wang, X., Chan, R. C., et al. (2025) · The Innovation Informatics · doi:10.59717/j.xinn-inform.2025.100011 · Cited by 5
Self-citationA General Framework for Efficient Medical Image Analysis via Shared Attention Vision Transformer
Liu, Y., Wen, Y., Yang, L., He, L., & Zhou, M. (2025) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2025.3644949 · Cited by 4
CAMP: continuous and adaptive learning model in pathology
Nguyen, A. T., Byeon, K., Kim, K., Song, B., Chae, S. W., & Kwak, J. T. (2026) · npj Artificial Intelligence · doi:10.1038/s44387-025-00051-5 · Cited by 10
“Among the four CAMP models (CAMP-CTransPath, CAMP-Phikon, CAMP-UNI, and CAMP-Virchow2),theperformanceofCAMP-CTransPathwas,ingeneral,inferior tothatoftheotherthreemodelsonbothpatch-andslide-levelclassi fi cation tasks.”
InfluentialHis‐MMDM: Multi‐Domain and Multi‐Omics Translation of Histopathological Images with Diffusion Models
Li, Z., Su, T., Zhang, B., Han, W., Zhang, S., Sun, G., Cong, Y., Chen, X., Qi, J., Wang, Y., et al. (2026) · Advanced Science · doi:10.1002/advs.202518066 · Cited by 7
Using unlabeled data to enhance fairness of medical AI
Movva, R., Koh, P. W., & Pierson, E. (2024) · Nature Medicine · doi:10.1038/s41591-024-02892-0 · Cited by 4
Lightweight SDENet Fusing Model-Based and Learned Features for Computational Histopathology
Bao, R., Zhao, Y., Srivastava, A., Frazier, S. R., & Palaniappan, K. (2024) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2024.3470312 · Cited by 4
“Model-based methods have not been well studied for medical image analysis tasks in microbiology [11] or computational pathology [12], [13], [14], [15], which is the focus of this current work.”
Cai, C., Li, J., Liu, M., Jiao, Y., & Xu, J. (2024) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm62325.2024.10822073 · Cited by 4
“…process of MIL can be expressed as follows. where f ( · ) denotes the feature extractor such as ImageNet pretrained ResNet [27], PLIP [23], UNI [24] and others; g ( · ) indicates the feature aggregation which aggregates and pools the extracted features to a single bag-level feature embedding;…”
Ye, Z., Wu, X., Wei, Z., Sun, Q., Wang, Y., Li, T., Yuan, Y., & Jing, J.-J. (2025) · Laboratory investigation; a journal of technical methods and pathology · doi:10.1016/j.labinv.2025.104101 · Cited by 4
A Multi‐Modal Approach for Exploring Sarcoma and Carcinoma Using FTIR and Polarimetric Analysis
Siddique, M. A., Akhtar, M., Majid, M. A., Khera, E. A., Ahmad, M., Gilanie, G., Hatamleh, W. A., Junaid, M. B., Dad, M. U., & Ullah, H. (2025) · Microscopy Research and Technique · doi:10.1002/jemt.24853 · Cited by 4
Lin, Q. D., Guan, S., Peng, M., Zhang, K., Zhang, H., Mo, T., & Yu, H. (2025) · Frontiers in Immunology · doi:10.3389/fimmu.2025.1513589 · Cited by 4
“In order to reduce the training time as well as to perform the dimensionality reduction of the data, we use the Transfer Learning and Convolutional Neural Networks approaches to transform each patch into vectors of size 1024 respectively using 2 pre-training models, ResNet50 and UNI (34, 35).”
Wong, B., Hong, S., & Yi, M. Y. (2025) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi60581.2025.10981015 · Cited by 7
“…practical guidance for feature extractor selection and may help explain the superior performance of current SOTA pathology foundation models [7, 15], which leverage deeper Transformer back-bones (e.g., Swin Tiny, ViT Large), advanced SSL techniques beyond basic contrastive learning, and…”
Exploring Foundation Models Fine-Tuning for Cytology Classification
Dausort, M., Godelaine, T., Zanella, M., Khoury, K. E., Salmon, I. J., & Macq, B. M. M. (2025) · IEEE International Symposium on Biomedical Imaging (ISBI) · doi:10.1109/isbi60581.2025.10981249 · Cited by 4
Lu, X.-J., Ying, Y., Chen, J., Chen, Z., Wu, Y., Prasanna, P., Chen, X., Jing, M., Liu, Z., & Lu, C. Y. (2025) · Brain‐X · doi:10.1002/brx2.70030 · Cited by 5
Streamlining tuberculosis detection with foundation model-based weakly supervised transformer
Bedőházi, Z., Biricz, A., Foster, N., Lin, Y. E., & Csabai, I. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.110554 · Cited by 4
STAN, a computational framework for inferring spatially informed transcription factor activity
Zhang, L., Sagan, A., Qin, B., Wang, H., Kim, E. S., Hu, B., & Osmanbeyoglu, H. U. (2025) · Nucleic Acids Research · doi:10.1093/nar/gkaf1473 · Cited by 4
“To address this limitation, we evaluated several pretrained deep learning models—TransPath [ 28 ], UNI [ 29 ], DenseNet-121 [ 30 ], and a bidirectional generative adversarial network (BiGAN) [ 31 ]— to extract more complex image features representing morphological structure ( Supplementary Methods…”
Histopathology-based Protein Multiplex Generation using Deep LearningPreprint
Andani, S., Chen, B., Ficek, J., Heinke, S., Casanova, R., Hild, B. F., Sobottka, B., Bodenmiller, B., Koelzer, V. H., & Rätsch, G. (2024) · medRxiv · doi:10.1101/2024.01.26.24301803 · Cited by 4
“We further explored the integration of features from the UNI computational pathology foundation model (FM) [43] to enhance the performance of HistoPlexer, marking one of the first attempts to apply pathology FMs to generative tasks.”
Self-citationAI-driven Discovery of Morphomolecular Signatures in ToxicologyPreprint
Jaume, G., Peeters, T., Song, A. H., Pettit, R. W., Williamson, D. F. K., Oldenburg, L., Vaidya, A. J., de Brot, S., Chen, R. J., Thiran, J., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.07.19.604355 · Cited by 4
Self-citationecPath detects ecDNA in tumors from histopathology imagesPreprint
Choudhury, M., Liu, L., Yadav, A., Chapman, O. S., Ahmadi, Z., Younis, R., Sharma, C., Goel, N., Sridhar, S., Kenkre, R., et al. (2024) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2024.11.13.623494 · Cited by 4
Multi-Modal Disentanglement of Spatial Transcriptomics and Histopathology ImagingPreprint
Maan, H., Ji, Z., Sicheri, E., Tan, T. J., Selega, A., Gonzalez, R., Krishnan, R. G., Wang, B., & Campbell, K. R. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.02.19.638201 · Cited by 4
“In our work, we use the UNI model (Chen et al., 2024) to obtain informative representations of the histology images ( X h ).”
InfluentialSharma, A., Liu, B., & Rantalainen, M. (2025) · Scientific Reports · doi:10.1038/s41598-025-04043-y · Cited by 3
“Each data split was stratified by the clinical NHG.”
The tumour histopathology “glossary” for AI developers
Mandal, S., Baker, A., Graham, T. A., & Bräutigam, K. (2025) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1012708 · Cited by 4
Shen, X., Zhang, F., Tang, C., Sokóvić, M., Mišić, D. M., Xu, H., Yang, Y., & Liu, J. (2025) · Rapid Communications in Mass Spectrometry · doi:10.1002/rcm.10045 · Cited by 4
“Notable examples include AI-based classification of rare cancers [98] and prediction of breast cancer patients’ responses to neoadjuvant chemotherapy [99].”
Investigation of Class Separability Within Object Detection Models in Histopathology
Ammeling, J., Ganz, J., Wilm, F., Breininger, K., & Aubreville, M. (2025) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2025.3560134 · Cited by 3
“In contrast, our method is not currently applicable to recent feature extractors like UNI [53] and Virchow [54], which are large transformer models trained on diverse data.”
AI-assisted Diagnosis of Nonmelanoma Skin Cancer in Resource-Limited Settings
Ellis, S., Song, S., Reiman, D. J., Hui, X., Zhang, R., Shahriar, H., Argos, M., Kamal, M., Shea, C. R., Grossman, R. L., et al. (2025) · Cancer Epidemiology Biomarkers & Prevention · doi:10.1158/1055-9965.epi-25-0132 · Cited by 3
“Yet UNI is worse than PRISM at distinguishing Bowen disease.”
InfluentialGao, R., Yuan, X., Sun, Y., Wang, Y., Xia, Y., Wei, T., Xu, D., & Yu, Z. (2025) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2025.108971 · Cited by 3
Scaling Chest X-Ray Foundation Models From Mixed Supervisions for Dense Prediction
Wang, F., & Yu, L. (2025) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2025.3589928 · Cited by 5
“These techniques have also shown promising results in specialized fields, such as radiology [27] and computational pathology [28].”
Decrypting cancer's spatial code: from single cells to tissue niches
Celik, C., Shi, P., Withnell, E., Lam, H. W., & Secrier, M. (2025) · Molecular Oncology · doi:10.1002/1878-0261.70100 · Cited by 4
Wang, C., Muzakky, H., Chung, Y.-P., Lai, P.-J., & Chao, T. (2025) · Medical Image Analysis · doi:10.1016/j.media.2025.103742 · Cited by 3
Historical Report Guided Bi-modal Concurrent Learning for Pathology Report Generation
Zhang, L., Yun, B., Li, Q., & Wang, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-04978-0_33 · Cited by 3
Ren, Q., Wang, Y., Fang, R., Ling, H., & You, C. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-05182-0_43 · Cited by 7
CytoSAE: Interpretable Cell Embeddings for Hematology
Dasdelen, M. F., Lim, H., Buck, M. C., Götze, K. S. S., Marr, C., & Schneider, S. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-05185-1_8 · Cited by 7
Carloni, G., Brattoli, B., Keum, S., Park, J., Lee, T., Ahn, C. H., & Pereira, S. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-07845-2_5 · Cited by 3
From Big to Small: Emerging Methods for Enhancing Precision Psychiatry Through Transfer Learning
Wang, Y.-W., Holmes, A. J., Yeo, B. T. T., & Yip, S. W. (2025) · Biological Psychiatry · doi:10.1016/j.biopsych.2025.10.022 · Cited by 3
Intratumour, anatomical and temporal heterogeneity in mesothelioma
Arnould, M., Neilly, M. D. J., Blyth, K. G., & Jean, D. (2025) · Thorax · doi:10.1136/thorax-2024-222052 · Cited by 3
Fiala, G., Plass, M., Harb, R., Regitnig, P., Skok, K., Zoughbi, W. A., Zerner, C., Torke, P. R., Kargl, M., Müller, H., et al. (2026) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2026.103368 · Cited by 3
Arlova, A., Maher, J. Y., Anvari, H., Badger, T. C., Hanfling, S. N., Balasubramanian, R., Lou, H., Sierra, M. D. L. L., Yun, B. H., Türkbey, B., et al. (2025) · Biology of Reproduction · doi:10.1093/biolre/ioaf250 · Cited by 4
Zabihollahy, F., Shi, K., Wangulu, C., Prassas, I., Masoomian, M., Saleeb, R., Gabril, M. Y., van der Kwast, T., Fleshner, N. E., & Yousef, G. M. (2025) · Modern Pathology · doi:10.1016/j.modpat.2025.100934 · Cited by 3
Fuhlert, P., Westhaeusser, F., Dietrich, E., Lennartz, M., Khatri, R., Kaiser, N., Röbeck, P., Bülow, R. D., von Stillfried, S. F., Witte, A., et al. (2025) · Medical Image Analysis · doi:10.1016/j.media.2025.103884 · Cited by 4
“Some examples of foundation models for histopathology are, among others, UNI (Chen et al., 2024), Midnight (Karasikov et al., 2025), or Virchow (Vorontsov et al., 2023).”
Cruciani, F., Nicoletti, G., Cafaro, D., Mauri, G., Lazzari, L., Aquilano, M. C., Regge, D., Marsoni, S., & Giannini, V. (2025) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc58623.2025.11253533 · Cited by 3
“Deep learning (DL) models have gained prominence in histopathological image analysis for feature extraction [3].”
Fu, M., Liu, L., Fu, F., Ouyang, J., Shi, X., Duan, S., Guan, T., He, Y., Cheng, Z., & Zhu, L. (2025) · Scientific Reports · doi:10.1038/s41598-025-19195-0 · Cited by 4
“As dataset scales expanded, the UNI model leveraged the DINO v2 algorithm with the ViT-L architecture, but it was trained on a large internal private dataset 21 .”
Zhang, H., Xu, H., & Lv, R. (2025) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2025.109063 · Cited by 3
Foundation models: Insights and implications for gastrointestinal cancer
Shi, L., Huang, R., Zhao, L.-L., & Guo, A.-J. (2025) · World Journal of Gastroenterology · doi:10.3748/wjg.v31.i47.112921 · Cited by 3
Wan, M., Ding, Y., Wang, Y., Jia, Y., Wu, S., Qu, W., Xu, Y., Fu, W., Timko, M. P., Wan, L., et al. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-02282-x · Cited by 4
“Paired with UNI — a self-supervised ViT pretrained on one of the largest histopathology slide collections — CLAM achieves strong classi fi cation performance (AUC 0.98 for LUAD vs LUSC; 0.97 for skin cancer) but has not been employed for survival modeling, with no applications reported in ICC 20 .”
InfluentialSelf-citationToward a robust and generalizable metamaterial foundation model
Kim, N., Lee, D., Yu, J., Cho, S. W., Lee, D., Park, Y., & Hong, Y. (2025) · npj Computational Materials · doi:10.1038/s41524-025-01925-7 · Cited by 3
“Foundation models have demonstrated remarkable success across a wide range of disciplines, including natural language processing, drug discovery 26 , pathology 27 , and molecular modeling 28 , by leveraging general-purpose architectures trained on vast datasets.”
Similarity-guided swarm of models: enhancing semi-supervised learning in computational pathology
Weng, Z., Pryalukhin, A., Hulla, W., Bychkov, A., Fukuoka, J., Schallenberg, S., Buchstab, O., Klauschen, F., Büttner, R., & Tolkach, Y. (2025) · Scientific Reports · doi:10.1038/s41598-025-33281-3 · Cited by 3
“Block-level feature embeddings were then computed using the foundational encoder UNI 35 , which is a state-of-the-art, general-purpose foundational self-supervised encoder built upon ViT-Large for pathology pre-trained on large amounts of histopathological data.”
InfluentialAn Introduction to Pathology Foundation Models
Veremis, B. M., Chen, S., & Campanella, G. S. V. (2026) · Head and Neck Pathology · doi:10.1007/s12105-025-01878-9 · Cited by 3
Wang, G., Chen, W., Liang, Z., Ding, F., Cui, J., Shen, Z., Lin, F., Li, X., Wang, X., & Zhao, X. (2026) · Communications Biology · doi:10.1038/s42003-026-09571-5 · Cited by 3
“The UNI model, a foundation model in computational pathology, is used to embed each patch into a 1024-dimensional feature vector 64 .”
Agrawal, A., & Thomann, S. (2026) · American Journal Of Pathology · doi:10.1016/j.ajpath.2026.01.007 · Cited by 4
Zhang, X., Wang, S., Gu, Q., Fu, Y., Li, H., Gan, J., Du, J., Chu, L., Li, X., Wang, C., et al. (2026) · Cancer Cell International · doi:10.1186/s12935-026-04211-8 · Cited by 4
“The feature extraction process utilized a pre-trained image encoder known as UNI, which is based on the Vision Transformer (ViT) architecture [14].”
Lalchungnunga, H., Dampier, C. H., Singh, O., Hoang, D.-T., Shulman, E. D., Abdullaev, Z., Li, B., Luo, Z., Wu, Z., Pearce, T. M., et al. (2026) · The Lancet Oncology · doi:10.1016/s1470-2045(25)00661-8 · Cited by 3
WSISum: WSI summarization via dual-level semantic reconstruction
Wang, B., Zhang, K., Wang, Y., Gu, Y., Luan, H., Zhou, Y., Hu, T., Wang, R., Yang, Z., Jiang, Z., et al. (2026) · Medical Image Analysis · doi:10.1016/j.media.2026.103970 · Cited by 3
Large Language Models and Otolaryngology
Bao, J. W., Jawad, M., Pavelchek, C., Powell, W. J. B., Oh, I. Y., Gupta, A., & Shew, M. A. (2026) · JAMA Otolaryngology–Head & Neck Surgery · doi:10.1001/jamaoto.2025.5335 · Cited by 3
Gindra, R. H., Palla, G., Nguyen, M., Wagner, S., Tran, M., Theis, F. J., Saur, D., Crawford, L. A., & Peng, T. (2025) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw69036.2025.00128 · Cited by 13
“Each input to the vision encoder is an image patch I i surrounding a Xenium patch, and the encoder outputs a feature vector v i 2 R d , where d is the feature dimension.”
InfluentialYue, J., Liu, J., Kang, X., Yuan, P., Wang, W., Wang, Z., Shang, C., Shang, Q., Li, G., Dong, X., et al. (2026) · Frontiers in Oncology · doi:10.3389/fonc.2026.1770037 · Cited by 3
“Through self-supervised and multitask learning, the UNI model captures complex morphological patterns and features from cancer histopathology, enabling ef fi cient and accurate feature extraction even on new datasets (32).”
Cano, F., Caviedes, M., Siabatto, A., Villarreal, J., Quijano, J. M., Bedoya-Urresta, Á., Bedoya, M., Caicedo, Y. Y., Roa, Á. A. C., González, F. A., et al. (2026) · Scientific Reports · doi:10.1038/s41598-025-32737-w · Cited by 4
“The UNI2-h foundational model, showed the poorest performance for intestinal metaplasia classification with F1-Score 0.72 ± 0.01 and AUC 0.84 ± 0.01 , for the internal test set, and F1-Score 0.62 ± 0.01 and AUC 0.75 ± 0.01 for the external test set.”
InfluentialTime and person sensitive foundation model for disease prediction and risk stratification
Wang, Z., Zhou, Y., Wu, Y., Goh, J. H. L., Zou, K., Guan, Z., Chen, Y., Yang, G. D., Zhang, P., Yin, C., et al. (2026) · npj Digital Medicine · doi:10.1038/s41746-026-02524-6 · Cited by 3
Learning with less supervision: A survey of label-efficient learning for medical image analysis
Jin, C., Guo, Z., Lin, Y., Luo, L., & Chen, H. (2026) · Medical Image Analysis · doi:10.1016/j.media.2026.104062 · Cited by 49
Alibrahim, M. N., Gloghini, A., & Carbone, A. (2026) · Blood Reviews · doi:10.1016/j.blre.2026.101390 · Cited by 3
Boosting pathology foundation models via few-shot prompt-tuning for rare cancer subtyping
He, D., Zhou, X., Guan, W., Zhang, L., Zhang, X., Xu, S., Wang, G., Wang, L., Yuan, X., Ma, J., et al. (2026) · Nature Communications · doi:10.1038/s41467-026-71715-2 · Cited by 7
“the development of pathology foundation models 11 – 23 .”
Gao, R., Yang, Z., Yuan, X., Wang, Y., Xia, Y., Zhang, Y., Zheng, B., Gong, Y., Yue, Y., Tu, Y., et al. (2026) · Nature Communications · doi:10.1038/s41467-026-71558-x · Cited by 6
“…pathological foundation models using different Meta-Encoder fusion strategies tailored for diverse downstream patch-level and WSI-level cancer detection tasks. features from individual patches, we selected three representative models published recently: CHIEF 20 , GigaPath 21 , and UNI 22 .”
DxDirector: an agentic large language model driving the full-process clinical diagnosis
Xu, S., Huang, X., Wei, Z., Pang, L., Shen, H., & Cheng, X. (2026) · Nature Communications · doi:10.1038/s41467-026-71928-5 · Cited by 7
“For instance, DxDirector-7B could call specialized pathology and imaging analysis models with visual understanding capabilities for Radiology 35 – 38 , Echocardiography 39 , Cell Slice Analysis 40 , Pathology 41 , and related domains.”
Explainable deep learning approaches and clinical insights for cancer biomarker identification
Srivastava, K., & Srivastava, R. (2026) · Frontiers in Oncology · doi:10.3389/fonc.2026.1810793 · Cited by 3
“Coupled with advances in foundation models for pathology and large language models for clinical data interpretation, these systems are moving toward fully automated, end-to-end precision oncology pipelines (114 – 116).”
Polit, F., Bahmad, H. F., Kassab, M. B., Elajami, M. K., Recine, M. A., Alghamdi, S., & Poppiti, R. (2026) · Frontiers in Oncology · doi:10.3389/fonc.2026.1833926 · Cited by 3
“…using DINOv2 self-supervised learning on over 100 million histopathology image patches from more than 100,000 whole slide images spanning 20 major tissue types, and demonstrated state-of-the-art performance across 34 clinical tasks including cancer subtyping, grading, and biomarker prediction (96).”
AI and the digital pathology revolution: clinical applications in cancer diagnosis and assessment
Paul, S., Isaacs, H. M., & Côté, R. J. (2026) · Expert Review of Molecular Diagnostics · doi:10.1080/14737159.2026.2665801 · Cited by 3
Shou, Y., Cao, X., Yan, P., Qiaohui, Zhao, Q., & Meng, D. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01853 · Cited by 9
“Since WSIs are generally very large, e.g., can be with 1 million by 1 million pixels, a common strategy in WSI-based survival analysis is to divide a WSI into patches sharing a same label, which can thus be formulated as a multi-instance learning (MIL) problem [1, 6, 29, 37].”
Spatial biomarker discovery via interpretable semantic learning in histopathology
Liang, J., Jiang, X., Reitsam, N. G., Lenz, T., Zhang, L., Gustav, M., Carrero, Z. I., Muti, H. S., Neidlinger, P., Clusmann, J., et al. (2026) · Cancer Cell · doi:10.1016/j.ccell.2026.05.014 · Cited by 3
End-to-end multimodal pathology foundation model with clinical dialogue
Vorontsov, E., Shaikovski, G., Casson, A., Viret, J., Zimmermann, E., Tenenholtz, N., Wang, Y. K., Bernhard, J., Godrich, R., Retámero, J. A., et al. (2026) · Nature Medicine · doi:10.1038/s41591-026-04521-4 · Cited by 25
“These models improved performance, robustness and data efficiency in cancer detection, subtyping and biomarker quantification 2,5–8 .”
Chen, K., Zhang, A., Xie, X., & Bai, Y. (2026) · International Dental Journal · doi:10.1016/j.identj.2026.109817 · Cited by 3
“11,12 Furthermore, as cloud-based platforms, these models eliminate the need for expensive local hardware, potentially lowering access barriers for advanced diagnostic assistance via simple web interfaces.”
Self-citationBatchkala, G., Li, B., & Rittscher, J. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73748-0_2 · Cited by 4
A foundation model unlocks unified biomedical image analysis
Huang, Y., Dou, H., & Ni, D. (2024) · Nature Methods · doi:10.1038/s41592-024-02519-9 · Cited by 3
An initiative on digital nephrology: the Kidney Imageomics Project
Zhou, F., Li, Z., Li, H., Lu, Y., Cheng, L., Zhang, Y., Wang, Z., Nie, J., Cheng, H., Dong, B., et al. (2025) · National Science Review · doi:10.1093/nsr/nwaf034 · Cited by 3
Multi-resolution Histopathology Patch Graphs for Ovarian Cancer Subtyping
Breen, J., Allen, K. E., Zucker, K., Orsi, N. M., & Ravikumar, N. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-83243-7_7 · Cited by 5
Metadata
- Authors
- 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., Williams, M., Oldenburg, L., Weishaupt, L. L., Wang, J. J., Vaidya, A., Le, L. P., Gerber, G., Sahai, S., Williams, W., & Mahmood, F.
- Affiliations
- Broad Institute (US)
- Brigham and Women's Hospital (US)
- Harvard University (US)
- Corresponding author
- Mahmood, Faisal
- ORCID
- Venue
- Nature Medicine
- Published
- 2024-03
- Volume
- 30
- Issue
- 3
- Pages
- 850-862
- ISSN
- 1078-8956 (ISSN-L)
- 1546-170X
- PMC
- PMC11403354
- OpenAlex
- W4392947521
- Semantic Scholar
- cac496ddebbff6bbba78242780cde01bd962bfc9
- Cited by
- 1,560
- References
- 170
- Citations by year
- 2026: 789 · 2025: 613 · 2024: 154
- FWCI
- 229.15
- Type
- journal article
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Licence
- Locations
- Nature Medicine (publishedVersion)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- Funders
- National Human Genome Research Institute (T32HG002295)
- National Cancer Institute (T32CA251062)
- National Institute of General Medical Sciences (R35GM149270)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- computational pathology, whole slide images, self-supervised learning, foundation models, unsupervised feature learning, tissue classification, few-shot learning, cancer subtyping, histopathology, transfer learning
- MeSH terms
- Artificial Intelligence, Workflow
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
