Automated Leukaemia Detection Using Microscopic Images

Abstract

In this paper, automated approach of leukaemia detection is proposed. In a manual method of Leukaemia detection, experts check the microscopic images. This is lengthy and time taking process which depends on the person's skill and not having a standard accuracy. The automated Leukaemia detection system analyses the microscopic image and overcomes these drawbacks. It extracts the required parts of the images and applies some filtering techniques. K-mean clustering approach is used for white blood cells detection. The histogram equalization and Zack algorithm is applied for grouping white blood cells. Some of the features like mean, standard deviation, colour, area, perimeter, etc. are calculated for leukaemia detection. The SVM is used for classification. The proposed system is tested on image dataset and 93.57% accuracy is achieved. The proposed system is successfully implemented in MATLAB.