Cloud based Early Acute Lymphoblastic Leukemia Detection Using Deep learning based Improved YOLO V4
Raj Kumar Gudivaka, Layth Hussein, T M Aruna, R Rana Veer Samara Sihman Bharattej, Priyan Malarvizhi Kumar
Abstract
Leukemia is fatal cancer in both children and adults and is divided into acute and chronic. Acute lymphoblastic leukemia (ALL) is a subtype of this cancer. Early diagnosis of this disease can have a significant impact on the treatment of this disease. However, recently many of the models are created for ALL detection, still the models are suffering from the detecting the tiny blood cell from the collected ALLIDB1 dataset. Proposed method is trained using Improved you only look once version four (YOLO v4) and is uploaded into Hadoop Distributed File System (HDFS) Hadoop framework. The collected dataset contains imbalance classes, so initially the dataset is undertaken by the preprocessing technique of data augmentation method for creating the synthetic data and solve the classes imbalance issue. The processed images are fed into the proposed YOLO v4 model to detecting the tiny blood cells. The suggested Improved YOLOv4 algorithm is shown to have a high effectiveness in enhancing the detection and recognition precision of healthy and blast cells.