Skip to main content

Deep Learning toolbox for WSI (digital histopatology) analysis

Project description

Codacy Badge PyPI version

DigiPathAI

A software application built on top of openslide for viewing whole slide images (WSI) and performing pathological analysis

Citation

If you find this reference implementation useful in your research, please consider citing:

@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}
}

Features

  • Responsive WSI image viewer
  • State of the art cancer AI pipeline to segment and display the cancerous tissue regions

Application Overview

Results

Online Demo

https://digipathai.tech/

Installation

Running of the AI pipeline requires a GPU and several deep learning modules. However, you can run just the UI as well.

Just the UI

Requirements

  • openslide
  • flask

The following command will install only the dependencies listed above.

pip install DigiPathAI

Entire AI pipeline

Requirements

  • pytorch
  • torchvision
  • opencv-python
  • imgaug
  • matplotlib
  • scikit-learn
  • scikit-image
  • tensorflow-gpu >=1.14,<2
  • pydensecrf
  • pandas
  • wget

The following command will install the dependencies mentioned

pip install "DigiPathAI[gpu]"

Both installation methods install the same package, just different dependencies. Even if you had installed using the earlier command, you can install the rest of the dependencies manually.

Usage

Local server

Traverse to the directory containing the openslide images and run the following command.

digipathai <host: localhost (default)> <port: 8080 (default)>

Python API usage

The application also has an API which can be used within python to perform the segmentation.

from DigiPathAI.Segmentation import getSegmentation

prediction = getSegmentation(img_path, 
			patch_size  = 256, 
			stride_size = 128,
			batch_size  = 32,
			quick       = True,
			tta_list    = None,
			crf         = False,
			save_path   = None,
			status      = None)

Contact

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

DigiPathAI-0.1.5.tar.gz (2.5 MB view details)

Uploaded Source

Built Distribution

DigiPathAI-0.1.5-py3-none-any.whl (2.6 MB view details)

Uploaded Python 3

File details

Details for the file DigiPathAI-0.1.5.tar.gz.

File metadata

  • Download URL: DigiPathAI-0.1.5.tar.gz
  • Upload date:
  • Size: 2.5 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/2.0.0 pkginfo/1.5.0.1 requests/2.23.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.44.1 CPython/3.7.7

File hashes

Hashes for DigiPathAI-0.1.5.tar.gz
Algorithm Hash digest
SHA256 997ecc517f72772157770b863000d869d20a9375287b533738f4a8c7ed9c5878
MD5 e05ca68b0da90f21ddac558845cd2d57
BLAKE2b-256 4988360733b5aaf3ff7d4156cc28ac3f578f065969db2ee55020471dd1f2bab6

See more details on using hashes here.

File details

Details for the file DigiPathAI-0.1.5-py3-none-any.whl.

File metadata

  • Download URL: DigiPathAI-0.1.5-py3-none-any.whl
  • Upload date:
  • Size: 2.6 MB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/2.0.0 pkginfo/1.5.0.1 requests/2.23.0 setuptools/40.8.0 requests-toolbelt/0.9.1 tqdm/4.44.1 CPython/3.7.7

File hashes

Hashes for DigiPathAI-0.1.5-py3-none-any.whl
Algorithm Hash digest
SHA256 6ba963c42a61028a773d4326c87949b542f87ee7c96c43802b4537c58a97b599
MD5 f614c4c8a1d38a14f869f788a2861955
BLAKE2b-256 51dea3e8d4009994293704d35eb7fc99e64a4fb468273a90134d302d8f41303f

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page