Medical-Image-Analysis

This repository contains work done on medical image analysis using AI, specifically lesion detection & segmentation of CT scans. Currently it contains work on analysis of DICOM images that are the most common means of working with medical images. Picked up from the Qure.AI CQ500 dataset (http://headctstudy.qure.ai/explore_data), these images have been worked on using Python and converted to numpy arrays for easy handling by ML models. The other work here is a simple lesion detection model constructed on the Head CT-hemorrhage dataset obtained from Kaggle (https://www.kaggle.com/felipekitamura/head-ct-hemorrhage) using Pytorch and associated libraries with python.