Comparison of classification techniques in the diagnosis of acute myelogenous leukemia

The leukemia blood cell discriminates from other blood cells by geometrical structure and cellular material by hematologist with microscope. In conventional color imaging system, the granular region in leukemia cells classify false negatively by experts due to micro geometrical and morphological changes in leukemia cell. Hence, a computer aided decision support system (CADSS) employs for automatic classification of leukemia blood cells. This paper gives the comparative study of various methods of classification of Acute Myelogenous Leukemia (AML) with improved accuracy and their results are compared. In this paper, the classification techniques such as Convolutional Neural Networks (CNN), Support Vector Machines (SVM), Fraction Black-Widow Optimized Neural Network (FB-NN), Deep CNN with Arithmetic Optimization and Hybrid Convolutional Bi-LSTM based RNN are made with blood microscopic images, in which according to measures, Deep CNN with Arithmetic Optimization has the best accuracy

Acute leukemia · Artificial neural network · Convolutional neural network · Deep learning · Hematologist · Leukemia · Pathology · Pattern recognition (psychology · Support vector machine · Computer Science · Digital Imaging for Blood Diseases · Medicine · Artificial Intelligence · Immunology