A New Method for Diagnosis and Predicting Blood Disorder and Cancer Using Artificial Intelligence (Artificial Neural Networks)

• Main Authors: Mehrdad Payandeh, Mehrnoush Aeinfar, Vahid Aeinfar, Mohsen Hayati
• Format: Article
• Language: English
• Published: Tehran University of Medical Sciences 2009-12-01
• Series: International Journal of Hematology-Oncology and Stem Cell Research
• Subjects: Artificial Neural Network; Multilayer Perceptron (MLP); Multi-Variable Non-linear Regression; Blood
• Online Access: https://ijhoscr.tums.ac.ir/index.php/ijhoscr/article/view/227

This paper represents a novel use of artificial neural networks in medical science. The proposed technique involves training a Multi Layer Perceptron (MLP) (a kind of artificial neural network) with a BP learning algorithm to recognize a pattern for the diagnosing and prediction of five blood disorders, through the results of blood tests from H1 machine. The blood test parameters and diagnosis of physician about the diseases of 450 patients from Taleghani Hospital in Kermanshah, Iran, are used in a supervised training method to update network parameters. This method was implemented to diagnose these disorder and cancer: Megaloblastic Anaemia, Thalassemia, Idiopathic thrombocytopenic pupura (ITP), Chronic myelogenous leukemia and Lymphoproliferative.