Skip to main content

​
Welcome to DigiPathAI

DigiPathAI is a comprehensive software application built on top of OpenSlide for viewing whole slide images (WSI) and performing AI-powered pathological analysis. It combines an intuitive web-based viewer with state-of-the-art deep learning models for automated tissue segmentation.

Interactive WSI Viewer

Responsive web-based interface for navigating and exploring whole slide images with support for multiple formats

AI-Powered Segmentation

State-of-the-art deep learning pipeline for automated cancer detection and tissue segmentation

Multi-Tissue Support

Pre-trained models for colon, liver, and breast tissue analysis with ensemble predictions

Python API

Programmatic access to segmentation pipeline for integration into custom workflows

​
What is DigiPathAI?

DigiPathAI provides a complete solution for digital pathology image analysis:
  • WSI Viewer: A browser-based interface powered by Flask and DeepZoom for seamless navigation of large whole slide images
  • AI Segmentation: Deep learning models (DenseNet, Inception, DeepLabv3) trained on multiple datasets for accurate tissue segmentation
  • Flexible Deployment: Run as a standalone viewer or with full GPU-accelerated AI capabilities
  • Open Source: Built on proven technologies like PyTorch, TensorFlow, and OpenSlide
DigiPathAI has been validated in peer-reviewed research. See the arXiv paper for technical details on the generalized deep learning framework.

​
Key Capabilities

​
Whole Slide Image Viewing

View and navigate gigapixel pathology images directly in your browser. DigiPathAI supports all formats compatible with OpenSlide, including:
  • TIFF/SVS (Aperio)
  • MRXS (Mirax)
  • NDPI (Hamamatsu)
  • And many more

​
AI-Powered Tissue Analysis

Perform automated segmentation with pre-trained models for:
  • Colon: DigestPath dataset models
  • Liver: PAIP challenge models
  • Breast: Camelyon challenge models
Models are automatically downloaded on first use and support ensemble predictions for improved accuracy.

​
Python Integration

Integrate segmentation into your analysis pipelines with a simple Python API. Process images programmatically with full control over patch size, batch size, test-time augmentation, and more.

​
Use Cases

  • Research: Analyze large cohorts of WSI images with consistent, automated segmentation
  • Education: Interactive viewer for teaching digital pathology concepts
  • Algorithm Development: Baseline segmentation models for comparison and benchmarking
  • Quality Control: Rapid review of tissue samples with AI assistance

​
Getting Started

Installation

Install DigiPathAI with UI-only or full AI pipeline support

Quick Start

Launch the server and run your first segmentation in minutes

​
Citation

If you use DigiPathAI in your research, please cite: 
@article{khened2020generalized,
  title={A Generalized Deep Learning Framework for Whole-Slide Image Segmentation and Analysis},
  author={Khened, Mahendra and Kori, Avinash and Rajkumar, Haran and Srinivasan, Balaji and Krishnamurthi, Ganapathy},
  journal={arXiv preprint arXiv:2001.00258},
  year={2020}
}

​
Community & Support

Build docs developers (and LLMs) love

Get started for freeTalk to us