Stroke detection from brain CT-images and its volume visualization

This study presents a comprehensive methodology for the detection, classification, and volumetric analysis of stroke lesions from computed tomography (CT) images. The approach encompasses several key stages: Skull stripping is performed to remove non-brain tissues, enhancing the accuracy of subsequent analyses. Stroke regions are identified by analysing the symmetry between the two hemispheres of the brain in CT images. Advanced segmentation algorithms are applied to delineate the region of interest (ROI) corresponding to the stroke-affected area. Texture features from the segmented ROI are extracted to capture the characteristics of the stroke lesion. The extracted features are input into classifiers such as Support Vector Machines (SVM) and K-Nearest Neighbors (K-NN) to categorize the type of stroke. For haemorrhagic strokes, the segmented regions from the CT image stacks are used to visualize and quantify the volume of the stroke. The methodology was validated using datasets from five patients, demonstrating its potential in aiding clinical diagnosis and treatment planning for stroke patients.