Computerised Decision Support System for Remote Health Monitoring: A Systematic Review

Abstract

BackgroundRemote health monitoring plays a major role in handling the critical situation of patients and avoiding death and also enhancing the quality of healthcare services. The effective real time monitoring with accurate decision has to be made in advance with the help of decision making system by continuously acquiring biosignals. ObjectivesThe main objective was to outline the research on remote patient health monitoring system that constitutes the multimodal biosignal acquisition system, thereby providing multi-label classification and clinical decision support system (CDSS). Methods and resultsA review was conducted with search terms such as multi-label classification, clinical decision support system, context-awareness and remote health monitoring. The study criteria included the randomized clinical trials evaluating the impact of efficient remote health monitoring system which incorporates CDSS for context-awareness systems by correlating several vital signs. From the total papers (n=52) which were included in the review, the major concentration of the review is multi-label classification (n=21, 40%). Further, this article included the review in context-awareness methods (n=5, 10%), clinical decision support systems (n=12, 23%), different means of biosignal acquisition and pre-processing (n=5, 10%), databases and software techniques for developing learning algorithms (n=3, 6%) and from general category (n=6, 12%). Several studies were effectively included which provides faster diagnosis for critically ill-patients. It is decisive for the critically ill-patients to be treated at the right time with proper and effective treatment which can be done efficiently using the CDSS and multi-label classification. The disease labels are classified as single and multi-labels where multi-label classification includes the disease labels for the correlated multiple vital signs and single label classification includes disease labels for individual vital signs. Further, on developing the logical learning model using multi-label classification, decision support system can be enhanced using context-awareness methods to predict the future vital signs, thereby providing an alert to the patients or doctors to take necessary actions. ConclusionThe proposed system includes the model that provides the correlations of several biosignals like electrocardiogram (ECG), peripheral capillary oxygen saturation (SPO2), body temperature and heartbeat, thereby identifying the critical situations and making the decisions using CDSS that helps in taking the necessary clinical interventions.