Key result
Support vector machine with feature selection achieves ~84% accuracy for two-class cardiovascular disease prediction.
Why the study?
Chronic diseases require continuous long-term health monitoring to prevent worsening health statuses and enable earlier diagnosis and treatment.
Absolute Event Rate: 84.48% vs 84.81%
Describes the architecture of a mobile health-monitoring system for continuous real-time transmission of physiological signals.
Mobile monitoring may expand chronic disease surveillance; leaves open whether it improves outcomes in randomized trials.
As the lifetime of human being gets longer, the problems of chronic diseases grow more. In order to make sure the health statuses of patients are not getting worse, they must be health-monitored continuously in a long term. In this paper, a mobile health-monitoring system is built for patients in place of traditional health-caring manners, which not only gives patients more free spaces, but also can save medical resources, diagnose and predict diseases earlier. In the procedures of health-caring in-house and emergency treatment, a series of vital sensors are combined by integrating sensor network and wireless/mobile network technology to continuously transmit physiological signals of patients to a medical center in a real time, and then doctors can monitor the health statuses of patients exactly, thereby proceeding with diagnosing, recovering, and treatments.
No takes yet. Share an insight, caveat, or question.
Yin‐Fu Huang (2018) studied Cardiovascular disease (n=303). Support vector machine with self-adaptive harmony search feature selection vs. Support vector machine without feature selection was evaluated on Classification accuracy for 2-class heart disease prediction. The cardiovascular disease prediction mechanism using a support vector machine with feature selection achieved 84.48% accuracy for two-class prediction and 61.76% for five-class prediction.
Synapse has enriched 2 closely related papers on similar clinical questions. Consider them for comparative context: