Key points are not available for this paper at this time.
Anemia and other blood disorders are serious global health issues affecting millions of individuals. These conditions, often triggered by insufficient hemoglobin or red blood cells, can manifest through symptoms like fatigue, weakness, and reduced immune function. When such disorders progress into advanced stages, they can compromise organ function and overall quality of life making early diagnosis especially critical. In recent years, as the value of prompt detection has become increasingly clear, artificial intelligence (AI) and autonomous diagnostic technologies have begun to take center stage in the medical community. Machine learning models excel at parsing complex datasets and generating accurate, rapid assessments, thus offering clinicians robust decision-support tools. Through these AI-driven methods, healthcare professionals can better interpret patients’ blood metrics and clinical indicators, enabling them to identify diseases at earlier stages and develop more effective treatment strategies. This study proposes a machine learning–based approach to classify various types of anemia and related blood disorders, including iron deficiency anemia, leukemia, and thrombocytopenia. We trained five contemporary algorithms Decision Tree (DT), Random Forest (RF), CatBoost, Gradient Boosting (GB), and XGBoost using critical blood parameters such as white and red blood cell counts, hemoglobin levels, and platelet counts. Notably, Gradient Boosting emerged as the most accurate model, achieving an impressive 99.19% accuracy rate. These findings underscore how AI-powered autonomous diagnostic systems have the potential to revolutionize hematology by facilitating earlier and more precise disease detection.
Çakmak et al. (Sat,) studied this question.