LEUKEMIA CANCER DETECTION USING DEEP LEARNING
International Research Journal of Modernization in Engineering Technology and Science · 2023
People of all ages, including children and adults, are susceptible to the deadly condition known as leukemia, which is a cancer-related illness.It is a leading cause of death in the world.It is especially related to White Blood Cells (WBC), which affect the bone marrow and/or blood and are followed by a rise in the number of immature lymphocytes.As a result, a timely and accurate cancer diagnosis is a crucial prerequisite for effective treatment that increases survival rates.Currently, this condition is diagnosed manually via a study of blood samples collected from microscopic pictures, which is frequently highly sluggish, time-consuming, and less precise.Additionally, leukemic cells appear and shape remarkably similar to normal cells under a microscope, making detection more challenging.RCNN -based deep learning has produced state-of-the-art algorithms for image classification challenges during recent years, however there is still room for improvement in terms of effectiveness, learning process, and performance.In order to diagnose Leukemia by examining the microscopic images of blood samples, we proposed the deep learning algorithms yolov5, yolov8, faster RCNN, and SSD in this research study.Incorporating squeeze and excitation learning, which iteratively executes calibration on channel-wise feature outputs by explicitly modelling channel interdependencies, the proposed deep learning architecture emphasizes the channel associations on all levels of feature representation.Incorporating the squeeze-and-excitation process also improves the ability to distinguish between leukemic and normal cells based on their features, helps to strategically reveal informative features of Leukemia cells while suppressing fewer valuable ones, As0a0result, deep learning algorithms are better able to represent features.We show how combining the squeeze and excite learning procedures in a deep learning0model can improve the computer's capability to recognize leukaemia from microscopic pictures based on patient blood samples.To solve the issue of insufficient data and improve their performance even further, a comprehensive set of experiments are carried out on both cropped cells and full-size microscopic images.