Classification of Abnormal Red Blood Cells Using Deep Learning

The most common and dangerous abnormality in red blood cells (RBCs) is shape aberration. The primary techniques for identifying and validating an anemic stage (shape abnormality) are measuring hemoglobin levels or doing a manual microscopic examination of peripheral blood smears. Under a microscope, manually detecting abnormal cells can be difficult and time-consuming. This leads to subpar outcomes, unnecessary medication that shortens patients’ lives, and technician eye fatigue. This program aims to increase the detection of abnormal red blood cells through image processing. The Decision-Tree technique is used as a machine-learning categorization technique. Ten abnormal red blood cells were consequently discovered and identified by the system. Images used in the system had already been collected from hospital patients. The machine altered and classed the image. Results show the names of the abnormal red blood cells that the system identified in the image.