News & Publications
All updates in one place
News
5/26/2026
HistAI's CellDX Data Hub has reached 300,000 whole slide images — a diverse collection of benign and malignant cases across various organs and indications, available for self-service exploration and purchase via the platform UI or AI agent skills.
5/13/2026
Train a deployable WSI classifier on a real, commercial-grade dataset in about 20 minutes for under $10
2/24/2026
HistAI has added 50,000+ new whole slide images to CellDX DataHub, bringing the total to 212,000+ slides, including 48,333 IHC slides — all available for immediate access via the platform and AI workflows.
2/2/2026
HistAI is proud to launch the Data Hub Skills, a revolutionary toolset compatible with Claude Code, Gemini, and CODEX agents, providing autonomous access to over 160,000 Whole Slide Images.
10/29/2025
HistAI and Protege Partner to Deliver One of the Largest Whole-Slide Pathology Datasets to AI Developers
10/10/2025
The largest curated dataset of more than 100K of WSIs available publicly
8/20/2025
The Hibou Family of Pathology Foundation Models, with over 1.5M downloads on Hugging Face, is now available in the Microsoft Azure AI Foundry Model Catalog.
6/3/2025
HistAI has released a landmark open-source dataset of 112,000 whole slide images and fully digitized pathology reports.
6/3/2025
HistAI has launched the SPIDER initiative at HIMSS, and released 5M high-quality patches and AI models for Skin, Colorectal, and Thorax.
6/10/2024
HistAI has introduced Hibou-L and Hibou-B, its first family of large vision foundation models in pathology trained on over 1.1M whole slide images.
12/12/2023
HistAI has launched CELLDX, a cross-platform AI-powered slide viewer that enables pathologists and researchers to analyze, collaborate, and automate routine tasks.
Publications
Alexey Pchelnikov, Aleksei Pchelnikov · arxiv · 5/11/2026
CellDX AI Autopilot lets users without machine learning expertise build, tune, and deploy whole slide image classifiers through conversational AI, cutting hyperparameter tuning costs by over 30x.
Dmitry Nechaev, Alexey Pchelnikov, Ekaterina Ivanova · arxiv · 5/17/2025
The HISTAI dataset is a large, open-access collection of over 100,000 multimodal whole slide images with rich metadata.
Dmitry Nechaev, Alexey Pchelnikov, Ekaterina Ivanova · arxiv · 4/7/2025
The SPIDER is the largest open-access, patch-level dataset spanning multiple organs with expert annotations and strong baseline models.
Dmitry Nechaev, Alexey Pchelnikov, Ekaterina Ivanova · arxiv · 8/20/2024
The Hibou family of vision transformers, pretrained on over 1 million whole slide images.