Application of a Feed Forward Neural Network for the Prediction of Differential Anemia
Main Article Content
Abstract
Anemia is one of the most prevalent ailments afflicting the least developed nations. It refers to the deficiency of Red Blood Cells (RBC) count in the blood. A comprehensive blood test is conducted in clinical laboratories, employing manual approaches to analyze patient samples. The methodology involves initial data processing, ANN design, and Final output processing. This study evaluates the pathological findings from 250 individuals, focusing on Complete Blood Count (CBC). As detailed in the following subsection, the sample is divided into two sets: one for training the ANN model and another for testing its performance. Learning data provides an accuracy of 88.0%, and validation and test datasets provide 87.5% and 86.5% accuracy respectively. The Feedforward Neural Network was observed to effectively detect cases of anemia, suggesting increased accuracy in the diagnosis.