AUTOMATIC DETECTION OF WHITE BLOOD CANCER FROM BONE MARROW MICROSCOPIC IMAGES
Keywords:
Blood Smear images, BCCD Set, YOLO, Tiny YOLO, VGG-16 Architecture, Bounding Boxes, Automatic identification and counting.Abstract
One of the important medical test to evaluate overall human health condition
is the Complete Blood Cell Count (CBC). Traditionally these blood cells were counted
manually by visual inspection or by using the haemocytometer along with some chemical
compounds and other scientific equipment’s which is a tedious and more time consuming
task. To avoid this problem, the proposed work here is the machine learning approach of
automatic identification and counting of blood cells (RBC’s, WBC’s and Platelets) using
‘You Only Look Once’ in short YOLO object detection and classification algorithm to
automatically identify and count the blood cells from the blood smear images. This YOLO
framework will be trained the modified Blood Cell Count Data set (BCCD) of blood smear
images to automatically identify and count the RBS’s, WBC’s and Platelets from the blood
smear images. Thus overall the computer aided system of counting and detection enables to
count the blood cells just in less than a second and this can be useful in the practical
applications.
