ADVANCEMENTS IN LEUKEMIA DIAGNOSIS: A COMPREHENSIVE REVIEW OF DEEP LEARNING APPROACHES FOR WHITE BLOOD CELL IMAGE ANALYSIS
S. Dhiravidaselvi, P. Deepan · 2024
Cancer of the blood is one of the most prevalent and perilous forms of the disease. The aberrant and unregulated synthesis of white blood cells (WBC) in the bone marrow is the root cause of leukaemia, a kind of blood cancer. Getting an early diagnosis of leukaemia increases the likelihood that cancer can be cured with the appropriate therapy. One method that can be used to identify leukaemia is the counting of the quantity of white and red blood cells (RBC). In the conventional approach, the counting of white blood cells and red blood cells is performed with the assistance of a piece of machinery called a haemocytometer. These examinations take a lot of time and are quite difficult to understand, which might lead to errors in categorization. Image processing of microscopic pictures, which is very affordable due to the uncomplicated nature of the circuitry, is another method that may be used to improve the identification of cancer cells. Researchers have found cancer cells, but because they didn't look at all of the criteria at the same time, such as picture enhancement, noise reduction, image recognition, and so on, their findings aren't as accurate as they might be. The goal and purpose of this research is to create a new system that is capable of taking into account all of the aspects that have been discussed thus far. Image acquisition, image pre-processing, picture segmentation, edge detection, and feature extraction are some of the steps that are included in the process. These phases are used to identify the presence of cancer as well as the stages at which cancer progresses. The paper contributes to the process of measuring the amount of white blood cells and red blood cells, as well as their average cell sizes and whether or not they are regular or irregular. This information may be used to determine whether or not a person has cancer