Cancer Cell Detection in Human Blood Samples using Microscopic Images : A Comprehensive Approach with CNN Classification

Abstract

Blood testing is now considered one of the most significant clinical exams. The features of a blood cell (volume, shape, and colour) can provide important information about a patient's health. Manual inspection, on the other hand, is time-consuming and necessitates a high level of technical understanding. As a result, automatic medical diagnosis technologies are required to assist clinicians in quickly and accurately identifying disorders. The primary goal of blood cell segmentation is to isolate defective/abnormal cells from a complex background and segment it into morphological components using image processing techniques like contrast enhancement, thresholding, morphological operations etc. The suggested technique utilized here minimizes noise and improves segmentation visually. All earlier approaches used various segmentation strategies, resulting in lower efficiency than the proposed method. This work can be implemented using MATLAB environment.