Detection of Blood-Related Diseases Using Artificial Neural Networks

Abstract

Advances in artificial intelligence have led to the emergence of intelligent systems and the development of tools that can help doctors diagnose and make decisions. This paper explains how artificial intelligence, for example artificial neural networks, can improve this field of diagnosis. The proposed technique involves the training of multilayer perceptron (a type of artificial neural network) with a reverse propagation training algorithm to diagnose and predict five blood disorders, through the results of a complete blood count test (CBC). The results showed the accuracy and the reliability of the proposed diagnosis system with sensitivity, specificity and accuracy reached 75.78%, 98.94% and 97.86% respectively.