Deep Learning Breakthroughs in Leukemia Diagnosis: A Review of White Blood Cell Image Analysis

One of the most common and deadly forms of cancer is blood cancer. White blood cells (WBC) are produced abnormally and uncontrollably in the bone marrow at the start of leukaemia, a blood cancer. An effective cancer cure is more likely when leukemia is identified and treated early. Red and white blood cell counts (RBC counts) are one way to determine leukemia. A haemocytometer is used in the traditional way to count the number of red blood cells and white blood cells. These tests are lengthy and complex, so there's a chance they won't be correctly classified. Image processing of microscopic photos is another possible tactic for enhancing cancer cell identification because of the circuitry's simplicity. Despite the fact that cancer cells have been identified, the researchers' inability to simultaneously take into account all relevant factors, such as image identification, noise reduction, and image enhancement, casts doubt on the accuracy of their findings. The primary objective of this study is to develop a new system that can incorporate all of the aforementioned elements. Image acquisition, pre-processing, picture segmentation, edge detection, and feature extraction are among the process's numerous phases. These steps can be used to determine whether cancer is present and how far along it is. We now know more about how to measure WBCs and RBCs in blood samples, as well as their typical sizes and regularity or irregularity. The diagnosis of cancer is one potential application for this data.