BiCNN-CML: Hybrid Deep Learning Approach for Chronic Myeloid Leukemia

chronic myeloid leukemia (CML) is an uncommon kind marked by genetic changes in early myeloid cell progenitors. The only efficient approaches for identifying leukemia are blood smear analysis, bone marrow aspiration, and biopsy. Because the tests mentioned above are time-consuming and expensive, leukemia diagnosis needs automation. Image processing is a low-cost and efficient way of identifying leukemia using stained blood microscope pictures. The major goal of this Research is to improve the accuracy of CML diagnosis. Recently, deep learning technologies have been employed in the medical business for early cancer detection and therapy prescription. Existing algorithms are exclusively concerned with image segmentation and feature extraction, and output may be provided as a consequence. In this paper proposed BiCNN-CML framework for predicting the CML. The datasets are collected and normalized using gray scale and morphological operation. The segmentation has done with FCM and PSO. The features are selected using GLCM and finally classification has done with Bi-LSTM with CNN algorithm. the experimental results has shown the classification metrics..