Deep learning based tumor segmentation from whole slide images

MAHENDRA KHENED

MAHENDRA KHENED

Chennai, Tamil Nadu

We develop fully convolutional neural network-based algorithms for segmentation of cancerous tissue regions from whole slide images. ...learn more

Project status: Concept

Artificial Intelligence

Intel Technologies
DevCloud, OpenVINO

Overview / Usage

The project aims at developing algorithms for automated detection of liver cancer in whole-slide images (WSIs). The liver is a visceral organ most often involved in the metastatic spread of cancer. For the best practice, early diagnosis of liver cancer is important but many people don't even know that they have hepatitis. Hepatocellular Carcinoma(HCC) represents about 90% of primary liver cancers and constitutes a major global health problem. This work would enable automating pathological work-flow and thereby aid pathologists in saving time.

Methodology / Approach

Fully convolution neural networks have been shown to produce state-of-art results for image segmentation tasks. Keras framework would be utilized for the development of training and inference pipeline.

Technologies Used

Tensorflow, Keras, GPU, CPU

Collaborators

There are no people to show.

Comments (0)