Main Article Content
Leukemia is the kind of cancer caused in bone marrow due to the immature leukocytes. The manual blood testing method is a slow process and required more time with less accurate detection results. Several methods are available for the detecting AML from the images of blood cell. The major motive of this work is for developing the efficient technique for leukemia detection. The proposed approach developed the enhanced Deep CNN with arithmetic optimization algorithm based leukemia detection from the peripheral blood cellimages.Here, AML is detected by the process namelypre-processing phase, segmentation phase, feature extraction phase, and classification phase. In the pre-processing phase, the unwanted noises and redundant data are removed. The nucleus as well as masked cell image is segmented by utilizing the modified distance regularized level set evolution (DRLSE) algorithm. Then the feature extraction as well as the classification is carried out by theDCNNfor the recognition of the normal images and also the AML infected images. The classification accuracy and the performance are improved by the arithmetic optimization algorithm. The analysis is carried out by utilizing the image from Munich AML Morphology Database and CPTAC-AML and performance is computed dependsupon the metrics such as accuracy, precision, sensitivity, and specificity. The experimental results revealed that the proposed DCNN-AOA classifier has achieved the highest accuracy of 99.82%. The performances of the proposed classifier are compared to other approachesand the analysis shows that this method provides better performance when compared to othermethods.