Fast Algorithm Based on Parallel Computing for Sample Entropy Calculation

Sample entropy is a widely used method for assessing the irregularity of physiological signals, but it has high computational complexity, which prevents its application time-sensitive scenes. To improve performance sample analysis continuous monitoring clinical data, fast algorithm based on OpenCL was proposed in this paper. an open standard supported by majority graphics processing unit (GPU) and operating systems. Based protocol, fast-parallel algorithm, OpenCLSampEn, calculation. A series 24-hour heartbeat data were to verify robustness algorithm. Experimental results showed that OpenCLSampEn exhibits great accelerating performance. With common parameters, can reduce execution time 1/75 base when signal length larger than 60,000. also different embedding dimensions, tolerance thresholds, scales In addition, R package provided GitHub. We significant improvement computation entropy. The broad utility facing challenge future rapid growth quantity signals.