Dataset Overview
Definition：Multi-Organ Pathology WSI Imaging Datasetis provided byChangsha Langhui Information Technology Co., Ltd.A large-scale, full-sliced digital image database. The DITC VL Hole Slide Microscopy Image standard is the main subject, with 5 million eligible physical tablets. The original Level-0 data, all pyramid levels, focal/optical channels and metadata are maintained. Overriding HeE dye is the main subject, IHC single/multi-labeled antibody layer, special dyes by type, multi-organ spectral coverage.
| Dataset Name | Multi-Organ Pathology WSI Imaging Dataset |
| Total Data Volume | 5,000,000 Slides (Slide Level) |
| Imaging Modality | Pathology Whole-Slide Digital Images (WSI) |
| Scan standards | Prioritize DICOM VL Whole Slide Microscopy Image；Native + OME-TIFF/BigTIFF 6.3.1 |
| Scan Magnification | 20x and above scan magnification, retaining Level-0 and all pyramid levels |
| Colour type | HE is the subject; IHC mono/multi-label antibody layer; special dye by type; frozen/paraffin layer |
| Covered Organs | Full multi-organ spectrum; excessive concentration on a single cancer type or single staining source is not permitted |
| Diagnostic label | Pathologically Confirmed Slides, TNM Staging, Immunohistochemistry Results |
| Source Institution | Pathology Department of Partner Grade-A Tertiary Hospitals |
| Use Cases | Pathology AI foundation model training, automated WSI analysis, tumor classification and grading, immunohistochemistry interpretation |
Delivery Specifications and Field Requirements
The dataset strictly follows the PATH-WSI unified delivery index requirements, ensuring data quality and traceability。
Delivery & Counting：One row corresponds to one complete WSI of a qualified physical slide；Re-scans / pyramids / multiple magnifications / multiple regions are not double-counted
Note of calibre:500,000 Cases (Slide Level) = WSIs Corresponding to 5 Million Qualified Physical Slides；Number of patients / cases / parts / paraffin blocks / physical slides / WSIs counted separately
Coverage：Covering all scannable tissue on the slide and the full slide area；Specimen site, sampling method, paraffin block / slide number, staining and focal plane must be traceable
Positives and Composition：Pathologically confirmed slides as the main body；Benign cases, normal adjacent tissue, inflammation and uncertain / consultation cases are listed separately according to task requirements
Raw Data Requirements：Scanner-native whole-field digital slides + interoperable files；Thumbnails, screenshots and ROI crops may not be used as a substitute for WSI
Cross-Data Association：Must be stably linked with the pathology diagnostic report；Slide-level / case-level / ROI / cell-level annotations must not be mixed
2026 Frontier AI Research Progress
Domain Review：In 2025-2026, pathology WSI AI entered the foundation model stage. mSTAR (Nature Communications 2025) and TITAN (Nature Medicine 2025) achieved excellent results across 97 benchmark tasks through multimodal self-supervised learning, and Nature 2026 reported a clinical-grade autonomous cytopathology platform。
mSTAR Multimodal Pathology Foundation Model
Nature Communications, 2025The integration of WSI, pathological reports, and gene expression is a three-module, using 26,169 slide-level model pairs, 32 cancer types, and over 116 million patch images.
Clinical-Grade Autonomous Cytopathology (Whole-Slide Edge Tomography)
Nature, 2026CYBO combines real-time full-slice optic imaging and marginalisation, achieving cell-level classification and group morphology analysis, and advancing clinical-level autonomous cellular pathology.
TITAN Multimodal WSI Foundation Model
Nature Medicine, 2025Generate pathological reports through 335,645 WSI visual self-supervised learning and visual-linguistic alignment, providing a basic model for pathological AI universalization.
HSFLA Optimizes Surgical Strategy for Lung Adenocarcinoma Diagnosis
npj Digital Medicine, 2026The hybrid-supervised deep learning model achieved an accuracy of 95.6% (manual 84.7%), the consistency between automatic annotation of infiltration areas and manual pixel-level annotation is 86.6%。
Prov-GigaPath Series
Nature 2024, arXiv 2026GigaPath-Flash, Gigatime, for efficient WSI reasoning, spatial proteomic predictions, and advanced the efficient pathological base model.
Annotation Workflow & Quality Control
Image Acquisition & De-identification
Standardized acquisition workflow: complete data is exported directly from the equipment, and patient identifying information (PHI) is removed before the data is stored。
Initial Annotation (Specialist Physician)
Attending physicians annotate case by case against the standards, including lesion localization, morphological description and disease term determination。
Review (Associate Chief Physician or Above)
Experts with the title of Associate Chief Physician or above review the initial annotation results item by item, correcting erroneous annotations and supplementing missing dimensions。
Consistency Assessment
10% of samples are randomly selected and independently annotated by 3 physicians to calculate Fleiss' Kappa; below 0.Dimensions scored 75 are flagged for rework。
License and Usage Agreement
Academic Research License
For universities and research institutions, supporting academic research related to medical AI。After signing the agreement, a de-identified data subset is provided; the source must be acknowledged：Langhui Technology DataAssetsAPI。
Commercial License
For medical device companies and AI diagnostics companies, supporting medical AI product R&D and medical device registration。Provides full datasets + custom annotation + incremental update services。
Data Compliance Statement
All images come from legally authorized sources; PHI fields are fully de-identified and contain no information that can directly identify an individual, in compliance with the Personal Information Protection Law and the Data Security Law。
Customized services
Supports extended requirements such as adding specific disease types, multimodal annotation expansion, and custom training of AI-assisted diagnostic models。
Quick Facts
AI Frontier Research
In 2025-2026, pathology WSI AI entered the foundation model stage. mSTAR (Nature Communications 2025) and TITAN (Nature Medicine 2025) achieved excellent results across 97 benchmark tasks through multimodal self-supervised learning, and Nature 2026 reported a clinical-grade autonomous cytopathology platform。