Detection and Classification of Acute Lymphocytic Leukemia
Pradeep Kumar Das, Priyanka Jadoun, Sukadev Meher
Abstract
The research work aims to develop an automated detection and classification method for acute lymphocytic leukemia (ALL). Extraction of lymphocytes is accomplished by the color based k-means clustering technique. Then, shape, texture, and color features are extracted from the segmented image. Gray-level co-occurrence matrix (GLCM) and gray-level run-length matrix (GLRLM) algorithms are used to extract the features of nucleus. Moreover, Principal component analysis (PCA) is applied for dimensional reduction. Finally, an SVM (support vector machine) with an RBF kernel is employed to classify WBCs. The proposed method yields promising results with 96.00% accuracy and 92.64% sensitivity.
Topics & Concepts
Artificial intelligencePattern recognition (psychology)Principal component analysisSupport vector machineComputer scienceCluster analysisKernel (algebra)Feature extractionGray levelContextual image classificationImage (mathematics)MathematicsCombinatoricsDigital Imaging for Blood DiseasesAI in cancer detectionSmart Agriculture and AI