リアルタイム呼吸性洞性不整脈抽出法を用いた作業負荷の制御

抄録

Respiratory sinus arrhythmia (RSA) is well known as a noninvasive index of cardiac vagal activity, which has an important role in maintaining homeostasis. In this study, we improved the signal processing of RSA in the respiratory phase domain for real-time analysis and used results of this signal processing to control the mental arithmetic workloads of subjects. Four healthy subjects performed two experiments : a self-speeded workload and a real-time task controlled workload. In these experiments, we obtained R—R intervals from electrocardiograms with a sampling frequency of 1 kHz, and respiratory information from elastic chest band with a sampling frequency of 100 Hz. Further, we placed an accelerometer on the larynx and sampled at the rate of 100 Hz. First, we conducted the self-speeded task experiment for 10 minutes. Subjects were required to do mental arithmetic at their best pace and to take two rest periods, approximately 1-minute each. The amplitude of RSA after averaging 24 respirations was lower when the subjects were working than when they were resting (P <0.001); we found that the amplitude of RSA reflects even short-term working or resting. This suggests the possibility of controlling the mental workload by using RSA amplitude data. Second, the average of rest and working RSA amplitudes was set as a threshold. If the average RSA amplitude for a subject's last 24 respirations was higher than the threshold, we assigned a mental arithmetic task at a pace of 20 questions per minute; if the average was lower, we stopped the task. Results of the experiment showed that the RSA amplitude was controlled well by changing the workload. In both experiments, the value of the RSA amplitude varied among individuals and within individuals, but the sensitivity to a task remained unchanged within individuals.