Leukemia detection using SegNet and faster region-based convolutional neural network

Prevention of cancer is mostly attained by surveillance of the transformation 
zones. White blood cells (WBCs) are established in the bone marrow and 
intemperate growth of WBC leads to leukemia. Hematologists examine the 
microscopic images in manual method for predicting leukemia, but it is very 
complex process and without any guaranteed for accurate. In this proposed 
study, deep learning techniques involved to segment and classify the three 
types of leukemia like acute lymphocytic leukemia (ALL), acute myeloid 
leukemia (AML) and chronic lymphocytic leukemia (CLL) using the 
BioGps dataset. The purpose of deep learning in medical science enhances 
the accuracy and precision of determining leukemia in early stages. In this 
study, introducing a sigmoid stretching (SS) in pixel enhancement for 
preprocessing; SegNet (St) is comfort to extract the structural features of the 
leukocytes and to segment the normal and blast cells for a clear 
classification; faster region-based convolutional neural network (faster RCNN) carried under the process of classification and optimization done by 
dragon fly algorithm. The result of this work achieves best accuracy related 
to the existing techniques of convolutional neural network (CNN) such as 
support vector machine (SVM), k-nearest neighbors (kNN) and Bayesian 
model. This study achieves the accuracy rate of 97%, precision rate of 94% 
and sensitivity rate of 90% respectively with low complexity.