Forecasting Diseases Through Microscopic Image Analysis by Utilizing BNN

Relying on the physical acuity of a haematologist, the current approach to diagnosing blood disorders is both time-consuming and prone to errors. To aid in clinical decision-making, there is a need for an automated visual image analysis system Early, accurate, and secure diagnosis of leukemia, a cancer that impacts the development of immature and abnormal white blood cells called blasts, is crucial for successful treatment and the patient's overall survival. While a blood smear test is typically used to diagnose by observing the white blood cells, machine learning techniques have also been identified as useful for diagnosing diseases like leukaemia. However, these techniques can have a significant misclassification error rate. To address this, a deep learning system can be used to categorize microscope pictures for the examination of white blood count. The WBC difference count system consists of two modules: the detection method and the classification model. The first module, called the detection module, is responsible for processing the raw bone marrow smear images. It identifies various components such as WBCs, red blood cells, platelet counts, and coloring impurities through image segmentation. The detection module then classifies the cells based on their features, such as size and shape. Once the cells are identified, they are passed on to the classification module. The classification process comprises of two stages, where the first stage involves filtering out non-relevant cells, such as crushed or degenerated cells, which are of no use for leukemia diagnosis. In the second stage, the quantifiable WBCs are differentiated using the back propagation network technique.