Monitoring System for Sickle Cell Disease Patients by Using Supervised Machine Learning

Recently, the need for a real-time healthcare monitoring system that able to offer remote and personal health-care services has increased. The patients with Sickle Cell Disease (SCD) require continuous services of testing, following-up and monitoring. Offering these services to patients smoothly at any time needs to integrated healthcare system. The recent development in information systems and technologies facilitate introducing such healthcare systems. This paper proposed an integrated system model, which offers the services of testing, following-up and monitoring patients with (SCD). The proposed system uses support vector machine SVM, which is supervised machine learning approach to analyze the collected data of a specific patient and takes the appropriate action such as send alert message to the healthcare staff. To perform the classification process, four methods are applied with SVM algorithm, which are Sequential Minimal Optimization SMO, Rules JRIP, Tree Decision Stump and Naive Bays for comparative analysis. In this paper, many experiments were implemented based on the four machine learning algorithms to determine patients of SCD from normal patients. The results were promising as they show 99% classifications were accurate when using SMO algorithm.