GLRLM based Feature Extraction for Acute Lymphoblastic Leukemia(ALL) Detection

Abstract

This paper proposes a gray level run length matrix (GLRLM) based feature extraction technique for the detection of Acute Lymphoblas-tic Leukemia (ALL). ALL could be a fatal hematopoietic ailment which might cause death if it's not treated at the early stage. The GLRL matrix is a method for extraction of statistical textural features from the nucleus of the lymphocyte image. The extracted features are then supplied to the Support Vector Machine (SVM) for classi_cation. The experiments are performed on an publicly available dataset ALL-IDB1. The accuracy of the proposed scheme is found to be 96.97% for SVM classifier