Robust Approach for Medical Data Classification and Deploying Self-Care Management System for Sickle Cell Disease

Intelligent systems and smart devices have played the major role in improving the healthcare organisation in terms of continuous tele-monitoring therapy and maintaining telemedicine management system for sickle cell disease. The biggest challenge facing majority of patients is the fact that there is still a lack of communication with healthcare professionals. Smart home (out-of hospital care) can raise personal self-sufficiency in association with living independently for longer as this disease is considered life-long condition. By using a self-care management system, we tend to improve patient welfare and mitigate patient illness before it gets worse over time, particularly with elderly people. This paper describes the state of the art in pervasive healthcare applications and the communication technologies that assist healthcare providers to offer better services for patients. This research proposes an alert system that could send immediate information to the medical consultants once detects serious condition from the collected data of the patient. Furthermore, the system is able to track various types of symptoms through mobile application in the purpose of obtaining support from medical specialists when it is required. A machine-learning algorithm was conducted to perform the classification process. Four experiments were carried out to classify sickle cell disease patients from normal patients using machine-learning algorithm in which 99.5984% classification accuracy was achieved using Multi-layer perceptron. Classification using Core Vector Regression, Hyper Pipes and Zero-Rule based algorithms achieved classification accuracy of 95.9839 %, 87.9518% and 70.6827 %, respectively.