[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-1622":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":15,"forks30d":15,"starsTrendScore":19,"compositeScore":20,"rankGlobal":10,"rankLanguage":10,"license":21,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":24,"hasPages":22,"topics":25,"createdAt":10,"pushedAt":10,"updatedAt":30,"readmeContent":31,"aiSummary":32,"trendingCount":15,"starSnapshotCount":15,"syncStatus":33,"lastSyncTime":34,"discoverSource":35},1622,"Awesome-AI4DigitalPathology","lingxitong\u002FAwesome-AI4DigitalPathology","lingxitong","A Curated List of Awesome Works in Computational Pathology, Aiming to Serve as a One-stop Resource for Researchers, Practitioners, and Enthusiasts Interested in Digital Pathology.","",null,"Python",235,18,10,0,7,9,25,21,3.84,"Apache License 2.0",false,"main",true,[26,27,28,29],"ai4health","aswsome-list","digital-pathology","foundation-models","2026-06-12 02:00:30","# Awesome-AI4DigitalPathology\n\n\u003Cdiv align=\"center\">\n\n\n[![Awesome](https:\u002F\u002Fcdn.rawgit.com\u002Fsindresorhus\u002Fawesome\u002Fd7305f38d29fed78fa85652e3a63e154dd8e8829\u002Fmedia\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002Fsindresorhus\u002Fawesome)\n[![Digital Pathology](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTopic-Digital%20Pathology-8A2BE2.svg)](#)\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-CC0_1.0-blue.svg)](LICENSE.txt)\n[![PRs Welcome](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPRs-welcome-brightgreen.svg)](CONTRIBUTING.md)\n\n**📜 A Curated List of Awesome Works in AI4DigitalPathology, Aiming to Serve as a One-stop Resource for Researchers, Practitioners, and Enthusiasts Interested in AI4DigitalPathology.**  \n*Focused on papers, benchmarks, datasets, and open-source repositories for modern digital pathology.*\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"assets\u002FACPATH-logo.png\" alt=\"Awesome World Models\" width=\"100%\" style=\"border-radius: 15px; box-shadow: 0 4px 24px rgba(0,0,0,.1); margin: 5px 0;\">\n\u003C\u002Fp>\n\n\n*Photo Credit: [Gemini-Nano-Banana🍌](https:\u002F\u002Faistudio.google.com\u002Fmodels\u002Fgemini-2-5-flash-image)*.\n\u003C\u002Fdiv>\n\n---\n\n## 🚩 News & Updates\n\n_Major updates and repository announcements are shown below._\n\n🚧 **[Ongoing] Repository Refocus** — This list is being rebuilt around **AI4DigitalPathology**, with the original awesome-list visual style preserved.\n\n💡 **[Ongoing] Contributions Welcome** — If you would like to add missing papers, repos, or benchmarks, feel free to open a PR.\n\n📌 **[Ongoing] Repository Support** — If this list helps your research, consider sharing the repository and citing it in your own awesome lists.\n\n---\n\n## Overview\n\n- 🎯 [Aim of the Project](#aim-of-the-project)\n- 📖 [Surveys, Reviews, and Perspectives](#surveys-reviews-and-perspectives)\n- 🖨️ [Digital Slide Scanners and File Formats](#digital-slide-scanners-and-file-formats)\n- 🗂️ [Datasets and Benchmarks](#datasets-and-benchmarks)\n- 🧩 [Multiple Instance Learning](#multiple-instance-learning)\n- 🌐 [Federated Learning in Computational Pathology](#federated-learning-in-computational-pathology)\n- 🤖 [Patch-Level Foundation Models](#patch-level-foundation-models)\n- 🖼️ [Slide-Level Foundation Models and Slide Encoders](#slide-level-foundation-models-and-slide-encoders)\n- 🧫 [Cytology and Cervical Cytology in Pathology AI](#cytology-and-cervical-cytology-in-pathology-ai)\n- 🎨 [Generative Models for Computational Pathology](#generative-models-for-computational-pathology)\n- 🧬 [Computational Pathology with Multi-Omics](#computational-pathology-with-multi-omics)\n- 💬 [Vision-Language Models and Pathology Agents](#vision-language-models-and-pathology-agents)\n- 🧱 [Dense Prediction in Computational Pathology](#dense-prediction-in-computational-pathology)\n- 🏥 [Clinical Tasks and Applications](#clinical-tasks-and-applications)\n- 🧭 [Pathology Image Registration and Spatial Alignment](#pathology-image-registration-and-spatial-alignment)\n- 🚀 [Resources, Toolkits, and Open-Source Projects](#resources-toolkits-and-open-source-projects)\n- 🔭 [Future Trends and Hot Topics](#future-trends-and-hot-topics)\n- 🙏 [Acknowledgements](#acknowledgements)\n- 📝 [Citation](#citation)\n\n---\n\n## Aim of the Project\n\nComputational pathology has rapidly evolved from handcrafted image analysis pipelines to **whole-slide learning**, **foundation models**, **multimodal pathology-language systems**, and **morphology-to-omics prediction**.  \nAt the same time, the literature has become fragmented across pathology, machine learning, computer vision, spatial biology, and multimodal AI.\n\nThis repository aims to:\n\n- 🔍 **Organize** representative papers, datasets, toolkits, and repositories in computational pathology\n- 🗺️ **Provide** a clean map of the field from classical WSI learning to modern foundation models\n- 🤝 **Bridge** communities working on digital pathology, multimodal medicine, spatial biology, and medical AI\n- 📚 **Serve** as a compact reading list for new researchers and a practical reference for experienced practitioners\n- 🚀 **Track** open-source progress in pathology AI, especially around benchmarks and reproducibility\n\n---\n\n## Surveys, Reviews, and Perspectives\n\n\u003Cem>Surveys, reviews, and perspectives that summarize the evolution, challenges, and future directions of computational pathology.\u003C\u002Fem>\n\n- **Computational Pathology: Challenges and Promises**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CMIG%202011-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fabs\u002Fpii\u002FS0895611111000383)\n- **Digital Pathology and Artificial Intelligence**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Lancet%20Oncol%202019-1f77b4.svg)](https:\u002F\u002Fwww.thelancet.com\u002Fjournals\u002Flanonc\u002Farticle\u002FPIIS1470-2045%2819%2930154-8\u002Fabstract)\n- **AI in Digital Pathology for Precision Oncology**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Rev%20Clin%20Oncol%202019-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41571-019-0252-y)\n- **Computational Pathology White Paper**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-J%20Pathol%202019-1f77b4.svg)](https:\u002F\u002Fpathsocjournals.onlinelibrary.wiley.com\u002Fdoi\u002F10.1002\u002Fpath.5331)\n- **Digital Pathology in Nephropathology**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Rev%20Nephrol%202020-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41581-020-0321-6)\n- **Artificial Intelligence and Computational Pathology**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Lab%20Invest%202021-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41374-020-00514-0)\n- **Digital Pathology in Translational Medicine**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Mod%20Pathol%202022-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41379-021-00919-2)\n- **AI in Computational Pathology of Cancer**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Ann%20Rev%20Cancer%20Biol%202023-1f77b4.svg)](https:\u002F\u002Fwww.annualreviews.org\u002Fcontent\u002Fjournals\u002F10.1146\u002Fannurev-cancerbio-061521-092038)\n- **Computational Pathology in 2030**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-EBioMedicine%202023-1f77b4.svg)](https:\u002F\u002Fwww.thelancet.com\u002Fjournals\u002Febiom\u002Farticle\u002FPIIS2352-3964%2822%2900609-0\u002Ffulltext)\n- **AI for Digital and Computational Pathology**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Rev%20Bioeng%202023-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs44222-023-00096-8)\n- **Applications of Digital Pathology in Cancer**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Ann%20Rev%20Cancer%20Biol%202024-1f77b4.svg)](https:\u002F\u002Fwww.annualreviews.org\u002Fcontent\u002Fjournals\u002F10.1146\u002Fannurev-cancerbio-062822-010523)\n- **Explainable AI for Precision Pathology**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Ann%20Rev%20Pathol%202024-1f77b4.svg)](https:\u002F\u002Fwww.annualreviews.org\u002Fcontent\u002Fjournals\u002F10.1146\u002Fannurev-pathmechdis-051222-113147)\n- **AI in Digital Pathology: Diagnostic Meta-analysis**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-npj%20Digit%20Med%202024-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41746-024-01106-8)\n- **Pathology in the Era of Generative AI**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Lancet%20Digit%20Health%202024-1f77b4.svg)](https:\u002F\u002Fwww.thelancet.com\u002Fjournals\u002Flandig\u002Farticle\u002FPIIS2589-7500%2824%2900157-2\u002Ffulltext)\n- **Artificial Intelligence in Pathology**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Ann%20Rev%20Biomed%20Data%20Sci%202025-1f77b4.svg)](https:\u002F\u002Fwww.annualreviews.org\u002Fcontent\u002Fjournals\u002F10.1146\u002Fannurev-biodatasci-103123-095814)\n- **AI and Digital Tools in Cancer Pathology**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Lancet%20Digit%20Health%202025-1f77b4.svg)](https:\u002F\u002Fwww.thelancet.com\u002Fjournals\u002Flandig\u002Farticle\u002FPIIS2589-7500%2825%2900115-3\u002Ffulltext)\n---\n\n## Digital Slide Scanners and File Formats\n\n\u003Cem>Digital slide scanners, image formats, and technical standards that support whole-slide imaging acquisition, storage, and interoperability.\u003C\u002Fem>\n\n- **OpenSlide** — open-source library for reading WSI formats across scanner vendors. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-JPI%202013-1f77b4.svg)](https:\u002F\u002Fpmc.ncbi.nlm.nih.gov\u002Farticles\u002FPMC3815078\u002F) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fopenslide\u002Fopenslide)\n- **opensdpc** — Python library for processing SDPC whole-slide images. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FWonderLandxD\u002Fopensdpc)\n- **Bio-Formats** — library for reading and converting microscopy formats. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Methods%202010-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fnmeth.1426) [![Website](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWebsite-OME-ffb6c1.svg)](https:\u002F\u002Fwww.openmicroscopy.org\u002Fbio-formats\u002F)\n- **DICOM WSI** — standard for digital pathology image interoperability. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-JDI%202018-1f77b4.svg)](https:\u002F\u002Flink.springer.com\u002Farticle\u002F10.1007\u002Fs10278-017-0034-3) [![Website](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWebsite-NEMA-ffb6c1.svg)](https:\u002F\u002Fdicom.nema.org\u002Fdicom\u002Fdicomwsi\u002F)\n---\n\n## Datasets and Benchmarks\n\n\u003Cem>Representative datasets and evaluation benchmarks for computational pathology.\u003C\u002Fem>\n\n- **TCGA** — multi-cancer WSIs + clinical\u002Fmolecular. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-GDC-orange.svg)](https:\u002F\u002Fportal.gdc.cancer.gov\u002F)\n- **CPTAC** — proteogenomic + histology cohorts. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-CPTAC-orange.svg)](https:\u002F\u002Fproteomics.cancer.gov\u002Fprograms\u002Fcptac)\n- **CAMELYON16** — lymph node metastasis detection. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-CAMELYON16-orange.svg)](https:\u002F\u002Fcamelyon16.grand-challenge.org\u002FData\u002F)\n- **CAMELYON17** — WSI and patient-level metastasis. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-CAMELYON17-orange.svg)](https:\u002F\u002Fcamelyon17.grand-challenge.org\u002F)\n- **PANDA** — prostate cancer grading benchmark. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-PANDA-orange.svg)](https:\u002F\u002Fpanda.grand-challenge.org\u002F) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FDIAGNijmegen\u002Fpanda-challenge)\n- **PatchCamelyon (PCam)** — patch metastasis classification. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-PCam-orange.svg)](https:\u002F\u002Fgithub.com\u002Fbasveeling\u002Fpcam)\n- **MHIST** — colorectal polyp classification. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-MHIST-orange.svg)](https:\u002F\u002Fbmirds.github.io\u002FMHIST\u002F)\n- **NCT-CRC-HE-100K** — 100k colorectal patches. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Zenodo-orange.svg)](https:\u002F\u002Fzenodo.org\u002Frecord\u002F1214456)\n- **BCNB** — breast cancer nodule and biomarker dataset. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Website-orange.svg)](https:\u002F\u002Fbupt-ai-cz.github.io\u002FBCNB\u002F)\n- **MUT-HET-RCC** — intra-tumor heterogeneity and mutation dataset. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Figshare-orange.svg)](https:\u002F\u002Fdoi.org\u002F10.25452\u002Ffigshare.plus.c.5983795)\n- **HER2-Tumor-ROIs** — annotated ROIs for HER2 scoring. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-TCIA-orange.svg)](https:\u002F\u002Fwww.cancerimagingarchive.net\u002Fcollection\u002Fher2-tumor-rois\u002F)\n- **EBRAINS** — ultra-high-resolution whole-slide brain mapping. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-EBRAINS-orange.svg)](https:\u002F\u002Fdoi.org\u002F10.25493\u002FWQ48-ZGX)\n- **VisioMel** — melanoma and lymph node metastasis dataset. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-DrivenData-orange.svg)](https:\u002F\u002Fwww.drivendata.org\u002Fcompetitions\u002F148\u002Fvisiomel-melanoma\u002F)\n- **IMP** — multi-institutional cervical and tissue data. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-INESC%20TEC-orange.svg)](https:\u002F\u002Frdm.inesctec.pt\u002Fdataset\u002Fnis-2023-008)\n- **Selected Cohorts** — CPTAC multi-cancer cohorts. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-TCIA-orange.svg)](https:\u002F\u002Fwww.cancerimagingarchive.net\u002Fcollections\u002F)\n- **AGGC2022** — large-scale prostate Gleason scoring. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Grand%20Challenge-orange.svg)](https:\u002F\u002Faggc22.grand-challenge.org\u002F)\n- **TIGER** — breast TIL segmentation and scoring. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Grand%20Challenge-orange.svg)](https:\u002F\u002Ftiger.grand-challenge.org\u002F) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FDIAGNijmegen\u002Ftiger-challenge-eval)\n- **GlaS** — colon gland instance segmentation. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202017-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841516301736) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Warwick-orange.svg)](https:\u002F\u002Fwarwick.ac.uk\u002Ffac\u002Fcross_fac\u002Ftia\u002Fdata\u002Fglascontest\u002F)\n- **TCGA-TIL Maps** — pan-cancer TIL spatial maps. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Cell%20Reports%202018-1f77b4.svg)](https:\u002F\u002Fwww.cell.com\u002Fcell-reports\u002Ffulltext\u002FS2211-1247(18)31438-5) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-TCIA-orange.svg)](https:\u002F\u002Fwww.cancerimagingarchive.net\u002Fanalysis-result\u002Ftil-maps\u002F)\n- **BACH** — breast cancer classification and segmentation. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202019-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841518301789) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Website-orange.svg)](https:\u002F\u002Ficiar2018-challenge.grand-challenge.org\u002F)\n- **MoNuSeg** — multi-organ nucleus segmentation. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-TMI%202020-1f77b4.svg)](https:\u002F\u002Fieeexplore.ieee.org\u002Fdocument\u002F8880654) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Grand%20Challenge-orange.svg)](https:\u002F\u002Fmonuseg.grand-challenge.org\u002F)\n- **SICAPv2** — prostate cancer Gleason grading. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-JBHI%202020-1f77b4.svg)](https:\u002F\u002Fieeexplore.ieee.org\u002Fdocument\u002F9144365) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Mendeley-orange.svg)](https:\u002F\u002Fdata.mendeley.com\u002Fdatasets\u002F9xxm58dvs3\u002F1)\n- **UniToPatho** — colon cancer with class imbalance and domain shift. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ECCV%202020-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2009.00650) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Zenodo-orange.svg)](https:\u002F\u002Fzenodo.org\u002Frecord\u002F3934241)\n- **NADT-Prostate** — prostate cancer with androgen-deprivation therapy. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-medRxiv%202020-6A5ACD.svg)](https:\u002F\u002Fwww.medrxiv.org\u002Fcontent\u002F10.1101\u002F2020.09.29.20199711v1)\n- **MoNuSAC2020** — multi-organ nuclei segmentation and classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-TMI%202021-1f77b4.svg)](https:\u002F\u002Fieeexplore.ieee.org\u002Fdocument\u002F9446924) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Grand%20Challenge-orange.svg)](https:\u002F\u002Fmonusac-2020.grand-challenge.org\u002F)\n- **Lizard** — large-scale colonic nuclei benchmark. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ICCV%202021-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fhtml\u002FGraham_Lizard_A_Large-Scale_Dataset_for_Colonic_Nuclear_Instance_Segmentation_and_Classification_ICCV_2021_paper.html)\n- **PAIP** — liver cancer segmentation and survival prediction. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202021-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841521000577) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Website-orange.svg)](https:\u002F\u002Fpaip2019.grand-challenge.org\u002F)\n- **TissueNet** — large-scale cell segmentation across modalities. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Methods%202021-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41592-021-01249-6) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-DeepCell-orange.svg)](https:\u002F\u002Fdatasets.deepcell.org\u002F)\n- **BRACS** — breast carcinoma subtyping. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Sci%20Data%202022-1f77b4.svg)](https:\u002F\u002Fpmc.ncbi.nlm.nih.gov\u002Farticles\u002FPMC9575967\u002F) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fhistocartography\u002Fhact-net)\n- **CoNIC** — colon nuclei identification and counting. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202022-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841522000755) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Zenodo-orange.svg)](https:\u002F\u002Fzenodo.org\u002Frecord\u002F6559981)\n- **OV-Bevacizumab** — multimodal ovarian cancer response dataset. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Sci%20Data%202022-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41597-022-01127-6)\n- **BCI** — H&E to IHC translation benchmark. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202022-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2022\u002Fhtml\u002FLiu_Translating_From_HE_to_IHC_A_New_Trajectory_for_Translational_CVPR_2022_paper.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fbupt-ai-cz\u002FBCI)\n- **EBHI-Seg** — digestive tumor segmentation. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Sci%20Data%202022-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41597-022-01435-y) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Figshare-orange.svg)](https:\u002F\u002Ffigshare.com\u002Farticles\u002Fdataset\u002FEBHI-Seg\u002F19602495)\n- **HEROHE** — HER2 status prediction from routine H&E. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202022-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841521002369) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Website-orange.svg)](https:\u002F\u002Fherohe.inesctec.pt\u002F)\n- **DigestPath** — colonoscopytissue segmentation. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202022-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841521003571) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Grand%20Challenge-orange.svg)](https:\u002F\u002Fdigestpath2019.grand-challenge.org\u002F)\n- **OCELOT** — cell detection with tissue context. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202023-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2023\u002Fhtml\u002FRyu_OCELOT_Overlapping_Cell_on_Tissue_Dataset_for_Histopathology_CVPR_2023_paper.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Flunit-io\u002Focelot-benchmark)\n- **Benchmarking SSL on Pathology** — SSL benchmarking across pathology datasets. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202023-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2023\u002Fhtml\u002FKang_Benchmarking_Self-Supervised_Learning_on_Diverse_Pathology_Datasets_CVPR_2023_paper.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Flunit-io\u002Fbenchmark-ssl-pathology)\n- **EVA** — evaluation framework for oncology foundation models. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MIDL%202024-d62728.svg)](https:\u002F\u002Fopenreview.net\u002Fforum?id=FNBQOPj18N) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fkaiko-ai\u002Feva)\n- **HEST-1k \u002F HEST-Benchmark** — wsi and spatial transcriptomics benchmark. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-NeurIPS%202024-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.16192) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-HuggingFace-orange.svg)](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FMahmoodLab\u002Fhest)\n- **PathMMU** — pathology large multimodal model benchmark. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202024-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.16355) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FPathMMU-Benchmark\u002FPathMMU) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-HuggingFace-orange.svg)](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fjamessyx\u002FPathMMU)\n- **KidRare** — pediatric rare tumor WSI dataset. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Commun%202026-1f77b4.svg)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41467-026-71715-2) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-HuggingFace-orange.svg)](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FFirehdx233\u002FKidRare)\n- **HISTAI** — open-access WSI resource with models. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202025-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.12120) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FHistAI\u002FHISTAI)\n- **PLISM Benchmark** — robustness benchmark for pathology foundation models. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202025-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.19674) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fowkin\u002Fplism-benchmark)\n- **PFM-DenseBench** — dense prediction benchmark for pathology foundation models. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202026-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.03887) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Flingxitong\u002FPFM_Segmentation)\n- **PathBench** — multi-task and multi-organ foundation-model benchmark. [![Website](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWebsite-Leaderboard-ffb6c1.svg)](https:\u002F\u002Fbirkhoffkiki.github.io\u002FPathBench\u002F)\n- **Patho-Bench** — standardized pathology foundation-model benchmark. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fmahmoodlab\u002FPatho-Bench)\n- **HistoBoard** — unified dashboard for pathology foundation model benchmarks. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fwearewaiv\u002Fhistoboard) [![Website](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWebsite-Leaderboard-ffb6c1.svg)](https:\u002F\u002Fwearewaiv.github.io\u002Fhistoboard\u002F)\n- **Stanford PathBench** — pathology foundation model benchmark and leaderboard. [![Website](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWebsite-Leaderboard-ffb6c1.svg)](https:\u002F\u002Fpathbench.stanford.edu\u002F)\n- **Sinai Benchmark** — tile-level SSL benchmark for pathology foundation models. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fsinai-computational-pathology\u002FSSL_tile_benchmarks)\n- **STAMP** — solid tumor associative modeling benchmark in pathology. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FKatherLab\u002FSTAMP-Benchmark)\n- **THUNDER** — benchmark for classification, calibration, robustness, and segmentation. [![Website](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWebsite-Leaderboard-ffb6c1.svg)](https:\u002F\u002Fmics-lab.github.io\u002Fthunder\u002F)\n- **PathoROB** — robustness benchmark for pathology foundation models. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fbifold-pathomics\u002FPathoROB)\n- **MindLab-DP\u002FDatasets** — practical collection of digital pathology datasets. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FMindLab-DP\u002FDatasets)\n- **TCGA Processing Pipeline for MIL** — WSI preprocessing for weak supervision. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fliupei101\u002FPipeline-Processing-TCGA-Slides-for-MIL)\n---\n\n## Multiple Instance Learning\n\n\u003Cem>Multiple instance learning methods for weakly supervised whole-slide image analysis.\u003C\u002Fem>\n\n- **ABMIL** — attention-based deep multiple instance learning. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202018-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.04712) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FAMLab-Amsterdam\u002FAttentionDeepMIL)\n- **Clinical-grade WSI** — large-scale weakly supervised WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Med%202019-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41591-019-0508-1)\n- **CAMEL** — weakly supervised WSI segmentation via class activation maps. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ICCV%202019-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fhtml\u002FLi_Camel_A_Weakly_Supervised_Learning_Framework_for_Histopathology_Image_Segmentation_ICCV_2019_paper.html)\n- **DeepAttnMISL** — multi-scale attention-guided MIL for WSI survival prediction. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202020-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841520301535) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Futa-smile\u002FDeepAttnMISL)\n- **CLAM** — clustering-constrained attention MIL for WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20BME%202021-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41551-020-00682-w) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fmahmoodlab\u002FCLAM)\n- **DSMIL** — dual-stream MIL for WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202021-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.08939) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fbinli123\u002Fdsmil-wsi)\n- **Patch-GCN** — graph-based context-aware WSI survival modeling. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202021-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.13048) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fmahmoodlab\u002FPatch-GCN)\n- **DT-MIL** — deformable transformer for MIL on histopathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202021-d62728.svg)](https:\u002F\u002Flink.springer.com\u002Fchapter\u002F10.1007\u002F978-3-030-87240-3_34) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fyfzon\u002FDT-MIL)\n- **SparseConvMIL** — sparse convolutional context-aware MIL. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-COMPAY%202021-d62728.svg)](https:\u002F\u002Fproceedings.mlr.press\u002Fv156\u002Flerousseau21a.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FMarvinLer\u002FSparseConvMIL)\n- **TransMIL** — correlated MIL with transformers. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-NeurIPS%202021-d62728.svg)](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2021\u002Fhash\u002F10c272d06794d3e5785d5e7c5356e9ff-Abstract.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fszc19990412\u002FTransMIL)\n- **ReMix** — general MIL data augmentation method for WSIs. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ICCV%202021-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fhtml\u002FYang_ReMix_Towards_Image_Mixup_for_Whole_Slide_Image_Classification_ICCV_2021_paper.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FJiawei-Yang\u002FReMix)\n- **HIPT** — hierarchical transformer for gigapixel pathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202022-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2022\u002Fhtml\u002FChen_Scaling_Vision_Transformers_to_Gigapixel_Images_via_Hierarchical_Self-Supervised_Learning_CVPR_2022_paper.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fmahmoodlab\u002FHIPT)\n- **DTFD-MIL** — double-tier feature distillation MIL. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202022-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2022\u002Fhtml\u002FZhang_DTFD-MIL_Double-Tier_Feature_Distillation_Multiple_Instance_Learning_for_Histopathology_Whole_CVPR_2022_paper.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fhrzhang1123\u002FDTFD-MIL)\n- **ZoomMIL** — differentiable zooming for MIL on whole-slide images. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ECCV%202022-d62728.svg)](https:\u002F\u002Fwww.ecva.net\u002Fpapers\u002Feccv_2022\u002Fpapers_ECCV\u002Fpapers\u002F136810689.pdf) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fhistocartography\u002Fzoommil)\n- **IBMIL** — intervention-based MIL for deconfounded WSI prediction. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202022-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2022\u002Fhtml\u002FLin_Interventional_Multi-Instance_Learning_with_Deconfounded_Instance-Level_Prediction_CVPR_2022_paper.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FTencentAILabHealthcare\u002FIBMIL)\n- **GTP** — graph-transformer fusion for WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-TMI%202022-1f77b4.svg)](https:\u002F\u002Fieeexplore.ieee.org\u002Fdocument\u002F9779215) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fvkola-lab\u002Ftmi2022)\n- **MHIM-MIL** — masked hard instance mining for WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ICCV%202023-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.15254) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FDearCaat\u002FMHIM-MIL)\n- **ILRA-MIL** — low-rank MIL for WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ICLR%202023-d62728.svg)](https:\u002F\u002Fopenreview.net\u002Fforum?id=8hH4Q3f8c2) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-MIL__BASELINE-green.svg)](https:\u002F\u002Fgithub.com\u002Flingxitong\u002FMIL_BASELINE)\n- **LNPL-MIL** — learning from noisy pseudo labels for WSI MIL. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ICCV%202023-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2023\u002Fhtml\u002FShao_LNPL-MIL_Learning_from_Noisy_Pseudo_Labels_for_Promoting_Multiple_Instance_ICCV_2023_paper.html)\n- **MILBooster** — boosting WSI classification via distribution and correlation. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ICCV%202023-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2023\u002Fhtml\u002FQu_Boosting_Whole_Slide_Image_Classification_from_the_Perspectives_of_Distribution_ICCV_2023_paper.html)\n- **PromptMIL** — prompting language-image models for pathology MIL. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202023-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.03362) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FZhenghui-Wu\u002FPromptMIL)\n- **S4MIL** — structured state space models for pathology MIL. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202023-d62728.svg)](https:\u002F\u002Fconferences.miccai.org\u002F2023\u002Fpapers\u002F622-Paper3313.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FMICS-Lab\u002Fs4_digital_pathology)\n- **WiKG** — whole-slide image as a knowledge graph. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202024-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.07719) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FWonderLandxD\u002FWiKG)\n- **CA-MIL** — context-aware MIL for WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ICLR%202024-d62728.svg)](https:\u002F\u002Fopenreview.net\u002Fforum?id=rzBskAEmoc) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Folgarithmics\u002FICLR_CAMIL)\n- **AC-MIL** — attention-challenging MIL. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ECCV%202024-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.05351) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fdazhangyu123\u002FACMIL)\n- **LongMIL** — long-contextual MIL for WSI analysis. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-NeurIPS%202024-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.14195) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Finvoker-LL\u002FLong-MIL)\n- **RRT-MIL** — feature re-embedding for WSI analysis. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202024-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2024\u002Fhtml\u002FTang_Feature_Re-Embedding_Towards_Foundation_Model-Level_Performance_in_Computational_Pathology_CVPR_2024_paper.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FDearCaat\u002FRRT-MIL)\n- **RetMIL** — retentive MIL for long histopathology sequences. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202024-d62728.svg)](https:\u002F\u002Fpapers.miccai.org\u002Fmiccai-2024\u002F657-Paper1723.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FHongbo-Chu\u002FRetMIL)\n- **MambaMIL** — Mamba-based long-sequence MIL. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202024-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.06800) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fisyangshu\u002FMambaMIL)\n- **cDP-MIL** — robust MIL via cascaded Dirichlet process. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ECCV%202024-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.11448) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FHKU-MedAI\u002FcDPMIL)\n- **ViLa-MIL** — dual-scale vision-language MIL for WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202024-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2024\u002Fhtml\u002FShi_ViLa-MIL_Dual-scale_Vision-Language_Multiple_Instance_Learning_for_Whole_Slide_Image_CVPR_2024_paper.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FJiangbo-Shi\u002FViLa-MIL)\n- **SI-MIL** — self-interpretable MIL for gigapixel histopathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202024-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2024\u002Fhtml\u002FBhattacharya_SI-MIL_Taming_Deep_MIL_for_Self-Interpretability_in_Gigapixel_Histopathology_CVPR_2024_paper.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fbmi-imaginelab\u002FSI-MIL)\n- **FG-VSI** — fine-grained visual-semantic WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202024-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2024\u002Fhtml\u002FLi_Generalizable_Whole_Slide_Image_Classification_with_Fine-Grained_Visual-Semantic_Interaction_CVPR_2024_paper.html)\n- **AMD-MIL** — agent aggregator with mask denoise for WSI analysis. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ACM%20MM%202024-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.11664) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fsigsminx\u002FAMD-MIL)\n- **SAM-MIL** — spatial contextual aware MIL with SAM guidance. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ACM%20MM%202024-d62728.svg)](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.1145\u002F3664647.3681534) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FFangHeng\u002FSAM-MIL)\n- **DGR-MIL** — diverse global representation learning for robust WSI MIL. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ECCV%202024-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.03575) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FChongQingNoSubway\u002FDGR-MIL)\n- **FR-MIL** — distribution re-calibration MIL with Transformer. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-TMI%202025-1f77b4.svg)](https:\u002F\u002Fpubmed.ncbi.nlm.nih.gov\u002F39163176\u002F) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FPhilipChicco\u002FFRMIL)\n- **HMIL** — hierarchical MIL for fine-grained WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202024-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.07660) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FChengJin-git\u002FHMIL)\n- **PseMix** — pseudo-bag mixup augmentation for MIL-based WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-TMI%202024-1f77b4.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.16180) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fliupei101\u002FPseMix)\n- **Lin-MIL** — linear attention MIL for scalable WSI analysis. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202025-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.13417) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fcharlotterchtr\u002FLin-MIL)\n- **PackMIL** — pack-based MIL training framework for pathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202025-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.12917) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FFangHeng\u002FPackMIL)\n- **Flow-MIL** — normalizing-flow latent feature space for WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ICCV%202025-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2025\u002Fhtml\u002FMa_Flow-MIL_Constructing_Highly-expressive_Latent_Feature_Space_For_Whole_Slide_Image_ICCV_2025_paper.html)\n- **MIL_BASELINE** — unified implementation hub for pathology MIL methods. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Flingxitong\u002FMIL_BASELINE)\n- **MIL-Lab** — standardized MIL library with pretrained slide models. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fmahmoodlab\u002FMIL-Lab)\n- **MIL Tutorial** — hands-on tutorial for pathology MIL pipelines. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fguillaumejaume\u002Fmil-tutorial)\n\n---\n## Federated Learning in Computational Pathology\n\n\u003Cem>Federated learning methods and privacy-preserving frameworks for collaborative computational pathology.\u003C\u002Fem>\n\n- **HistoFL** — federated learning for WSI classification and survival prediction. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202022-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841521003431) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fmahmoodlab\u002FHistoFL)\n- **FedStain** — federated stain normalization for pathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202022-d62728.svg)](https:\u002F\u002Flink.springer.com\u002Fchapter\u002F10.1007\u002F978-3-031-16434-7_2) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FMECLabTUDA\u002FBottleGAN)\n- **FLamby** — cross-silo federated learning benchmark. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-NeurIPS%202022-d62728.svg)](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2022\u002Fhash\u002F5e637100c63800cc078ad0da3d1697e9-Abstract-Datasets_and_Benchmarks.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fowkin\u002FFLamby)\n- **FedCamelyon16** — federated Camelyon16 benchmark in FLamby. [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-FLamby-orange.svg)](https:\u002F\u002Fgithub.com\u002Fowkin\u002FFLamby\u002Ftree\u002Fmain\u002Fflamby\u002Fdatasets\u002Ffed_camelyon16)\n- **WSI-FL Tool** — federated training tool for WSI segmentation. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-JPI%202022-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS2153353922006952) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FSarderLab\u002Ffederated_learning)\n- **FedMM** — federated multimodal learning for computational pathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202024-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.15858)\n- **CPath-FL Review** — review of federated learning in computational pathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CSBJ%202024-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS200103702400357X)\n- **FLCP Review** — literature review of federated learning for computational pathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-JMI%202025-1f77b4.svg)](https:\u002F\u002Fwww.spiedigitallibrary.org\u002Fjournals\u002Fjournal-of-medical-imaging\u002Fvolume-12\u002Fissue-06\u002F061412\u002FFederated-learning-in-computational-pathology-a-literature-review\u002F10.1117\u002F1.JMI.12.6.061412.full)\n- **HistoFS** — non-IID WSI classification via federated style transfer. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202025-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2025\u002Fhtml\u002FRaswa_HistoFS_Non-IID_Histopathologic_Whole_Slide_Image_Classification_via_Federated_Style_CVPR_2025_paper.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Flalakitchen\u002FHistoFS)\n- **PathFL** — federated pathology image segmentation across centers. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202025-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841525002178) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fyuanzhang7\u002FPathFL)\n- **RW-CPath-FL** — real-world federated learning for clinical pathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-JPI%202025-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS2153353925000501)\n- **FedPathHarmony** — federated harmonization for multi-center pathology data. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fcollaborativebioinformatics\u002FFedPathHarmony)\n- **FedWSIDD** — federated WSI classification via dataset distillation. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202025-d62728.svg)](https:\u002F\u002Fpapers.miccai.org\u002Fmiccai-2025\u002F0331-Paper1647.html)\n- **Fed-cSCC** — federated model for cSCC progression prediction. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-npj%20Precis%20Oncol%202025-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41698-025-00997-4)\n- **FedDMIL** — federated deep MIL for histopathology WSI classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-BSPC%202026-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fabs\u002Fpii\u002FS1746809425015708)\n\n---\n\n## Patch-Level Foundation Models\n\n\u003Cem>Patch-level foundation models and visual encoders for histopathology representation learning.\u003C\u002Fem>\n\n- **CTransPath** — transformer-based self-supervised pathology encoder. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202022-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fabs\u002Fpii\u002FS1361841522002043) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fjamesdolezal\u002FCTransPath)\n- **HIPT** — hierarchical transformer for pathology images. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202022-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.02647) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fmahmoodlab\u002FHIPT)\n- **RetCCL** — contrastive pathology patch representation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202023-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fabs\u002Fpii\u002FS1361841522002730) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fjamesdolezal\u002FRetCCL)\n- **Lunit-DINO** — self-supervised ViT for pathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202023-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.04690) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Flunit-io\u002Fbenchmark-ssl-pathology)\n- **Phikon** — large-scale self-supervised pathology ViT. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-NeurIPS%202023-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.16864) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fowkin\u002Fphikon)\n- **PLIP** — pathology vision-language pretraining model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Med%202023-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41591-023-02504-3) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fvinid\u002Fplip)\n- **PathoDuet** — pathology foundation model for H&E and IHC. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202023-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.09894) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fopenmedlab\u002Fpathoduet)\n- **CONCH** — caption-based pathology foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Med%202024-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41591-024-02856-4) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FMahmoodLab\u002FCONCH)\n- **UNI** — general-purpose pathology foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Med%202024-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41591-024-02857-3) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FMahmoodLab\u002FUNI)\n- **Virchow** — clinical-grade pathology foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Med%202024-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41591-024-03141-0) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fpaige-ai\u002FVirchow)\n- **Virchow2** — mixed-magnification pathology encoder. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202024-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.00738) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fpaige-ai\u002FVirchow2)\n- **Phikon-v2** — upgraded pathology foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202024-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2409.09173) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fowkin\u002Fphikon-v2)\n- **Hibou** — DINOv2-based pathology vision transformer. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202024-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.05074) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fhistai\u002Fhibou-b)\n- **kaiko Pathology FMs** — large-scale pathology ViT family. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202024-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.15217) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002F1aurent\u002Fkaikoai-models)\n- **Prov-GigaPath** — pathology tile-level foundation encoder. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nature%202024-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07441-w) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fprov-gigapath\u002Fprov-gigapath)\n- **PLUTO** — lightweight multi-scale pathology foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202024-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.07905)\n- **UNI2-h** — second-generation pathology encoder from UNI. [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FMahmoodLab\u002FUNI2-h)\n- **H-Optimus-0** — open foundation model for histology. [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fbioptimus\u002FH-optimus-0)\n- **H-Optimus-1** — next-generation histology encoder. [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fbioptimus\u002FH-optimus-1)\n- **Path Foundation** — Google pathology patch encoder. [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fgoogle\u002Fpath-foundation)\n- **BEPH** — BEiT-based pathology foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Commun%202025-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41467-025-57587-y) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FZhcyoung\u002FBEPH)\n- **MUSK** — multimodal pathology foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nature%202025-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-08437-2) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fxiangjx\u002Fmusk)\n- **Digepath** — gastrointestinal pathology foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202025-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2505.21928) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fxtxx\u002FDigepath)\n- **PathOrchestra** — pathology foundation model for clinical tasks. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202025-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.24345) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fyanfang-research\u002FPathOrchestra)\n- **PLUTO-4** — next-generation PLUTO model family. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202025-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.02826)\n- **StainNet** — pathology foundation model for special stains. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202025-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2512.10326) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FJWonderLand\u002FStainNet)\n- **Midnight** — efficient pathology foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202025-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.05186) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fkaiko-ai\u002Fmidnight)\n- **OpenMidnight** — open reproduction of Midnight. [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FSophontAI\u002FOpenMidnight)\n- **GPFM** — pathology foundation model toolkit. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fbirkhoffkiki\u002FGPFM)\n- **KEEP** — knowledge-enhanced pathology vision-language model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Cancer%20Cell%202026-1f77b4.svg)](https:\u002F\u002Fwww.cell.com\u002Fcancer-cell\u002Ffulltext\u002FS1535-6108(26)00058-9) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FAstaxanthin\u002FKEEP)\n- **GenBio-PathFM** — pathology foundation model from public data. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Preprint%202026-6A5ACD.svg)](https:\u002F\u002Fgenbio.ai\u002Fpapers\u002Fgenbio-pathfm.pdf) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fgenbio-ai\u002Fgenbio-pathfm)\n- **Atlas 2** — clinical pathology foundation model family. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202026-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2601.05148) [![Website](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWebsite-Aignostics-ffb6c1.svg)](https:\u002F\u002Fwww.aignostics.com\u002Fproducts\u002Ffoundation-models)\n- **GloPath** — entity-centric renal pathology foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Advanced%20Science%202026-1f77b4.svg)](https:\u002F\u002Fadvanced.onlinelibrary.wiley.com\u002Fdoi\u002F10.1002\u002Fadvs.202520580)\n- **CerS-Path** — cervical subspecialty pathology foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202026-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.10196)\n\n\n---\n\n## Slide-Level Foundation Models and Slide Encoders\n\n\u003Cem>Slide-level foundation models and whole-slide encoders for gigapixel pathology image understanding.\u003C\u002Fem>\n\n- **Prov-GigaPath** — first whole-slide foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nature%202024-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07441-w) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fprov-gigapath\u002Fprov-gigapath)\n- **CHIEF** — clinical histopathology imaging evaluation foundation. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nature%202024-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-024-07894-z) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fhms-dbmi\u002FCHIEF)\n- **PANTHER** — morphological prototyping for slide foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202024-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.11643) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fmahmoodlab\u002FPANTHER)\n- **TANGLE** — transcriptomics-guided slide representation learning. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202024-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.11618) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fmahmoodlab\u002FTANGLE)\n- **PRISM** — multimodal foundation model for slide-level histopathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202024-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.10254) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fpaige-ai\u002FPrism)\n- **CPath-Omni** — unified multimodal foundation model spanning patches and WSIs. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202025-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2025\u002Fhtml\u002FSun_CPath-Omni_A_Unified_Multimodal_Foundation_Model_for_Patch_and_Whole_CVPR_2025_paper.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FPathFoundation\u002FCPath-Omni)\n- **SlideChat** — slide-level vision-language assistant model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-CVPR%202025-d62728.svg)](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2025\u002Fhtml\u002FChen_SlideChat_A_Large_Vision-Language_Assistant_for_Whole-Slide_Pathology_Image_Understanding_CVPR_2025_paper.html) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FGeneral-Medical-AI\u002FSlideChat_Weight)\n- **MADELEINE** — multistain pretraining for slide representation learning. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ECCV%202024-d62728.svg)](https:\u002F\u002Flink.springer.com\u002Fchapter\u002F10.1007\u002F978-3-031-73414-4_2) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FMahmoodLab\u002Fmadeleine)\n- **PathAlign** — vision-language model for whole-slide images in histopathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MIDL%202024-d62728.svg)](https:\u002F\u002Fproceedings.mlr.press\u002Fv254\u002Fahmed24a.html)\n- **TITAN** — multimodal whole-slide foundation model for pathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Med%202025-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41591-025-03982-3) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FMahmoodLab\u002FTITAN)\n- **THREADS** — molecular-driven foundation model for oncologic pathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202025-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.16652)\n- **FEATHER** — lightweight supervised slide foundation models. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ICML%202025-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.09960) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002FMahmoodLab\u002Ffeather)\n- **mSTAR** — knowledge-enhanced whole-slide foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Commun%202025-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41467-025-66220-x) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FWangyh\u002FmSTAR)\n- **EXAONE Path 2.5** — pathology foundation model with multi-omics alignment. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202025-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2512.14019) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FLGAI-EXAONE\u002FEXAONE-Path-2.5)\n- **WSI-Concepts** — supervised foundation model trained from whole-slide images. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202025-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.05742) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FFraunhoferMEVIS\u002FMedicalMultitaskModeling)\n- **HistoGPT** — slide foundation model for WSI report generation. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Commun%202025-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41467-025-60014-x) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002Fmarr-peng-lab\u002Fhistogpt)\n- **Democratizing_WSI \u002F GigaSSL** — optimized slide-level representations for TCGA-scale analysis. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Ftrislaz\u002FDemocratizing_WSI)\n- **MOOZY** — patient-first foundation model for computational pathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202026-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.27048) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FAtlasAnalyticsLab\u002FMOOZY)\n- **CARE** — molecular-guided slide-level foundation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202026-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2602.21637) [![Model](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FModel-HuggingFace-yellow.svg)](https:\u002F\u002Fhuggingface.co\u002FZipper-1\u002FCARE)\n\n---\n\n## Cytology and Cervical Cytology in Pathology AI\n\n\u003Cem>Cytology and cervical cytology studies for cell-level screening, diagnosis, and pathology AI applications.\u003C\u002Fem>\n\n- **Computational Cytology Survey**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202023-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS136184152200319X)\n- **Cervical Cytology Deep Learning Review**. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Neurocomputing%202024-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS0925231224014012)\n- **DeepPap** — deep learning for cervical cytology cell classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-JBHI%202017-1f77b4.svg)](https:\u002F\u002Fpubmed.ncbi.nlm.nih.gov\u002F28541229\u002F)\n- **Multi-Task Feature Fusion** — feature-fusion model for cervical cell classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-JBHI%202022-1f77b4.svg)](https:\u002F\u002Fpubmed.ncbi.nlm.nih.gov\u002F35671309\u002F)\n- **TDCC-Net** — task decomposing and cell comparing for cervical lesion cell detection. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-TMI%202022-1f77b4.svg)](https:\u002F\u002Fieeexplore.ieee.org\u002Fdocument\u002F9744114)\n- **Comparison Detector** — comparison-based detector for cervical cells. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Neurocomputing%202021-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fabs\u002Fpii\u002FS092523122100014X) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fkuku-sichuan\u002FComparisonDetector)\n- **Robust Cervical Detection** — local-scale consistency distillation for cell detection. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202023-d62728.svg)](https:\u002F\u002Fconferences.miccai.org\u002F2023\u002Fpapers\u002F552-Paper2082.html) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Ffeimanman\u002FCervical-Abnormal-Cell-Detection)\n- **CellGAN** — conditional cervical cell synthesis for data augmentation. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202023-d62728.svg)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-43987-2_47) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FZhenrongShen\u002FCellGAN)\n- **SIPaKMeD** — Pap smear dataset for cervical cell classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ICIP%202018-d62728.svg)](https:\u002F\u002Fwww.cse.uoi.gr\u002F~cnikou\u002FPublications\u002FC072%20-%20Plissiti%20-%20icip%202018%20-%20Athens.pdf) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Kaggle-orange.svg)](https:\u002F\u002Fwww.kaggle.com\u002Fdatasets\u002Fmarinaeplissiti\u002Fsipakmed)\n- **HiCervix** — hierarchical dataset and benchmark for cervical cytology classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-TMI%202024-1f77b4.svg)](https:\u002F\u002Fpubmed.ncbi.nlm.nih.gov\u002F38923481\u002F) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FScu-sen\u002FHiCervix)\n- **BMT** — cross-validated ThinPrep Pap dataset. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Sci%20Data%202024-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41597-024-04328-3)\n- **HMCHH-TCT-CellDet** — large ThinPrep cytologic test dataset. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Sci%20Data%202025-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41597-025-04374-5) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Figshare-orange.svg)](https:\u002F\u002Fspringernature.figshare.com\u002Farticles\u002Fdataset\u002FA_large_annotated_cervical_cytology_images_dataset_for_AI_models_to_aid_cervical_cancer_screening\u002F27901206)\n- **AIATBS** — AI-assisted TBS classification for cervical smears. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Commun%202021-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41467-021-23913-3)\n- **Cervical WSI Screening** — WSI analysis for cervical cancer screening. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Commun%202021-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41467-021-25296-x) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FShenghuaCheng\u002FAided-Diagnosis-System-for-Cervical-Cancer-Screening)\n- **Detection-Free Pipeline** — detection-free cervical WSI screening. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MICCAI%202023-d62728.svg)](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002F10.1007\u002F978-3-031-43987-2_24) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fthebestannie\u002FDetection-free-MICCAI2023)\n- **LESS** — label-efficient multi-scale learning for cytology WSIs. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202024-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841524000343) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fubc-tea\u002FLESS-WSI)\n- **STRIDE** — large-scale AI-assisted cervical cytology screening. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202024-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.19512)\n- **AICCS** — AI system for cervical cytology screening. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Commun%202024-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41467-024-48705-3) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fcellsvision\u002FAICCS)\n- **Patch-to-Sample Reasoning** — patch-to-sample reasoning for cervical WSI screening. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-TAI%202024-1f77b4.svg)](https:\u002F\u002Fwww.computer.org\u002Fcsdl\u002Fjournal\u002Fai\u002F2024\u002F06\u002F10285382\u002F1Rd2FdKWJHO)\n- **Smart-CCS** — cervical screening with pretraining and test-time adaptation. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-arXiv%202025-6A5ACD.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.09662) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fhjiangaz\u002FSmart-CCS)\n- **DualCytoNet** — AI-assisted cervical cytology for low-resource settings. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Commun%202025-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41467-025-62589-x)\n- **LBC-DL** — LBC model for cervical precancer and cancer detection. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Commun%202025-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41467-025-58883-3) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FLuZWCHA\u002FLBC_WSI_Classification)\n- **UniCAS** — foundation model for cervical cytology screening. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Cell%20Rep%20Med%202025-1f77b4.svg)](https:\u002F\u002Fwww.cell.com\u002Fcell-reports-medicine\u002Ffulltext\u002FS2666-3791%2825%2900643-3) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fpeter-fei\u002FUniCAS)\n- **CellProfiler** — open-source cell image analysis platform. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Methods%202012-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fnmeth.2083) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FCellProfiler\u002FCellProfiler)\n- **HoVer-Net** — nuclear instance segmentation and classification. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202019-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841519301045) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fvqdang\u002Fhover_net)\n- **MoNuSeg** — multi-organ nucleus segmentation benchmark. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-TMI%202019-1f77b4.svg)](https:\u002F\u002Fpmc.ncbi.nlm.nih.gov\u002Farticles\u002FPMC10439521\u002F) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Grand%20Challenge-orange.svg)](https:\u002F\u002Fmonuseg.grand-challenge.org\u002F)\n- **Cellpose** — generalist cellular segmentation model. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Methods%202021-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41592-020-01018-x) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FMouseLand\u002Fcellpose)\n- **Mesmer** — whole-cell and nuclear segmentation for multiplexed images. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Biotechnol%202021-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41587-021-01094-0) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fvanvalenlab\u002Fdeepcell-tf)\n- **Lizard** — large-scale colonic nuclei dataset. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ICCV%20Workshop%202021-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2108.11195) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Warwick-orange.svg)](https:\u002F\u002Fwarwick.ac.uk\u002Ffac\u002Fcross_fac\u002Ftia\u002Fdata\u002Flizard\u002F)\n- **MoNuSAC** — multi-organ nuclei segmentation and classification challenge. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-TMI%202021-1f77b4.svg)](https:\u002F\u002Fieeexplore.ieee.org\u002Fdocument\u002F9446924) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Grand%20Challenge-orange.svg)](https:\u002F\u002Fmonusac-2020.grand-challenge.org\u002F)\n- **PanNuke** — pan-cancer nuclei dataset. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-ECCV%20Workshop%202021-d62728.svg)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.10778) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Website-orange.svg)](https:\u002F\u002Fjgamper.github.io\u002FPanNukeDataset\u002F)\n- **NuCLS** — crowdsourced nuclei classification and segmentation dataset. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-GigaScience%202022-1f77b4.svg)](https:\u002F\u002Facademic.oup.com\u002Fgigascience\u002Farticle\u002Fdoi\u002F10.1093\u002Fgigascience\u002Fgiac037\u002F6586817) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Website-orange.svg)](https:\u002F\u002Fsites.google.com\u002Fview\u002Fnucls\u002Fhome)\n- **CoNIC** — colon nuclei segmentation and classification challenge. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202023-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841523003079) [![Dataset](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDataset-Grand%20Challenge-orange.svg)](https:\u002F\u002Fconic-challenge.grand-challenge.org\u002F)\n- **CellViT** — ViT for nuclei instance segmentation in pathology. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-MedIA%202024-1f77b4.svg)](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS1361841524000689) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FTIO-IKIM\u002FCellViT)\n- **CellViT++** — WSI-scale cell segmentation and classification framework. [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FTIO-IKIM\u002FCellViT-plus-plus)\n\n---\n\n\n## Computational Pathology with Multi-Omics\n\n\u003Cem>Computational pathology studies integrating histology with genomics, transcriptomics, proteomics, and other omics data.\u003C\u002Fem>\n\n\n- **DeepPATH** — histology-based cancer gene mutation prediction. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Med%202018-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41591-018-0177-5) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fncoudray\u002FDeepPATH)\n- **SpaCell** — morphology and ST to predict disease cells. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Bioinformatics%202020-1f77b4.svg)](https:\u002F\u002Facademic.oup.com\u002Fbioinformatics\u002Farticle\u002F36\u002F7\u002F2293\u002F5663455) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002FBiomedicalMachineLearning\u002FSpaCell)\n- **HE2RNA** — bulk RNA-seq prediction from WSIs. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20Commun%202020-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41467-020-17678-4) [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCode-GitHub-green.svg)](https:\u002F\u002Fgithub.com\u002Fowkin\u002FHE2RNA_code)\n- **ST-Net** — histology and ST for spatial gene expression. [![Paper](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPaper-Nat%20BME%202020-1f77b4.svg)](https:\u002F\u002Fwww.nature.com\u002Farticle","Awesome-AI4DigitalPathology 是一个专注于计算病理学领域的精选资源列表，旨在为研究人员、从业者和爱好者提供一站式的信息资源。该项目涵盖了论文、基准测试、数据集以及开源项目等核心内容，主要使用 Python 语言进行开发。它特别强调了数字病理学中的多实例学习、联邦学习、基础模型等多个技术方向。适用于需要深入了解或应用人工智能于数字病理学的场景，如医学研究、临床诊断支持及教育等领域。",2,"2026-06-11 02:45:03","CREATED_QUERY"]