Study of AI-Based Solutions for Automatic Detection of Some Diseases Related to Red Blood Cells in West Africa

Abstract

The majority of hematology laboratories in the West Africa does not have equipment dedicated to the automatic classification of blood cells. The integration of artificial intelligence (AI) in hematology improves diagnostic accuracy, reduces the burden on healthcare systems, and provide timely interventions in regions with limited access to medical resources. This paper discusses the development and implementation of AI-based tools designed to automatically detect diseases related to red blood cells (RBC) in West Africa. These tools leverage advanced machine learning algorithms to analyze blood cell morphology and identify abnormalities indicative of diseases such as sickle cell anemia, elliptocytosis and other blood disorders. An analysis of previous techniques shows that models based on artificial neural networks (ANNs) and convolutional neural networks (CNNs) are the best systems for automatically detecting pathologies, with performance over 80%. When these models are combined with classifiers such as support vector machine (SVM) and k-nearest neighbor (KNN), they achieve better performance, with values between 91% and 98%.