Biomedical Image Segmentation

The project is centered around analysis of human cells particularly cancer cells found in histopathological images for upstream therapeutic benefit. The pipeline is to be incorporated into the patient centric platform at  https://tison.lums.edu.pk/ 

Project Title

Project Area of Specialization

Artificial Intelligence

Project Summary

Project Objectives

Major objective is to provide an image analysis pipeline for histopathological images which are routinely conducted in cancer hospitals for cancer patients

Project Implementation Method

Techniques from computer vision and image processing will be applied to perform analysis of histopathological images which will integrated with the system at https://tison.lums.edu.pk/ and allow an interface to process results

Benefits of the Project

The benefits of artificial intelligence have the potential to transform the healthcare system in pakistan which is still under serving its people. Hospital staff can easily perform image analysis through our proposed project by utilizing the web interface at https://tison.lums.edu.pk/

Technical Details of Final Deliverable

A convolutional neural network will be trained on a large number of histopathology images taken from the Cancer Imaging Archive (https://www.cancerimagingarchive.net/) which will allow the network to predict and segment cellular regions in the images and image processing technqiues will be used to provide various metrics like cellular count, density, area amongnst other metrics