Blood Smear morphological analysis and optimization through machine learning model

This article examines the utility of image processing and machine learning in analyzing the shape and structure of cell pictures obtained from peripheral blood smears, particularly on automatic identification. The fundamental principles of the three primary components (segmentation, quantitative characteristics, and classification) are delineated, and contemporary works are examined. This study examines malignant lymphoids and normal cells. These technologies can assist cytologists in conducting practical, objective, and rapid morphological studies of blood cells. They can help analyze specific morphological characteristics and function as educational and research instruments. Further study is required to establish screening strategies that effectively utilize image-based automated recognition systems in testing facilities, alongside other analysis approaches, as part of their regular operations.