Classification of Leukemia White Blood Cell Cancer

Leukemia, a pervasive blood cancer, poses a significant global health challenge. This research study presents a comprehensive system for the classification of leukemia white blood cell cancer, utilizing advanced Image Processing and Machine Learning techniques implemented in MATLAB. Achieving a remarkable accuracy rate of 98%, the system employs the K-Nearest Neighbors (KNN) algorithm to analyze distinctive features extracted from leukemia cell images. The workflow encompasses key stages including Image Capture, Data Enhancement, Cell Isolation, Feature Identification, and Categorization. Preprocessing techniques such as normalization, contrast enhancement, and noise removal enhance the quality of input data, while segmentation methods reveal cancer cell nucleus regions with precision. Feature extraction encompasses statistical color features and texture features, facilitating precise classification into leukemia cell types such as ALL-L1, ALL-L2, AML-M2, and AML-M5. Additionally, the system provides audio outputs corresponding to identified leukemia cancer types, enhancing interpretability and usability. This paper plays a pivotal role in advancing precise leukemia diagnosis by emphasizing the potential of integrated image processing and machine learning methodologies to address critical healthcare challenges.