Anemia Diagnosis And Prediction Based On Machine Learning

Abstract

The extraordinary developments in the health sector have resulted in the substantial production of data in daily life. To get valuable information out of this data—information that can be used for analysis, forecasting, making suggestions, and making decisions—it must be processed. Accessible data is converted into useful information using data mining and machine learning approaches. The first challenge for medical practitioners in developing a preventative strategy and successful treatment plan is the timely diagnosis of diseases. Sometimes, this can result in death if accuracy is lacking. In this study, we examine supervised machine learning methods (Decision Tree, Multilayer Perceptron “MLP”, K-nearest neighbors “KNN”, Logistic Regression, Random Forest, and Support Vector Machine “SVC”) for anemia prediction utilizing CBC (Complete Blood Count) data gathered from pathology labs. The outcomes demonstrate that the Random Forest, Multilayer Perceptron “MLP”, Decision Tree, and Logistic Regression techniques outperform KNN and SVC in terms of accuracy of 99.94%.