A Non-Invasive Sleep Analysis Approach Based on a Fuzzy Inference System and a Finite State Machine

Abstract

This paper deals with the analysis of sleep quality, which involves a non-invasive sleep stage detection method with home deployability. Some physiological signals, such as heart rate, heart rate variation, and the number of times the subject rolled over, are collected to determine the sleep stage. A fuzzy inference system is adopted to evaluate the division of sleep stage. Then, a preliminary sleep depth is calculated. Furthermore, a finite-state machine is developed to detect the sleep stage changes. The difference between our research and other existing studies is that, first, both the pressure sensors and the heart rate device are employed; then, the fuzzy inference and a finite-state machine are introduced, which give us a higher precision than the traditional methods to evaluate the sleep stage. The experimental results show that the proposed method can well evaluate the sleep quality that is almost consistent with a polysomnography test. The latter is currently recognized as the best way to measure sleep quality, which, however, requires a variety of monitoring sensors and has to be performed by the nursing staff in a professional setting. The reported approach can be used for monitoring sleep quality or sleep disorder screening at home. INDEX TERMS Sleep analysis, non-invasive, fuzzy inference system, finite state machine.

Applications of non‐invasive sensor devices to personalised health care

“…Our early work provided a theoretical framework [18]; some technologies and implements are improved in this paper. As shown in Fig.…”

“…The percentage value of each column in Table 4 is the corresponding sleep time divided by the total sleep time. More subjects and their analysis can refer to our previous study [18] if readers are interested. Polysomnography (PSG) is a preferred method for monitoring sleep quality because it is considered the gold standard for objectively evaluating sleep.…”

“…In the previous research [18], we have discussed the fuzzy inference for the sleep stage. A method of extracting the heart rate, heart rate variation and the number of turning over during sleep by using a heartbeat belt and pressure sensor is presented.…”

“…A method of extracting the heart rate, heart rate variation and the number of turning over during sleep by using a heartbeat belt and pressure sensor is presented. We improve the method in [18] and directly transmit parameters using Bluetooth Low Energy (BLE) technology. We have actually implemented the technology to generate a product.…”

The Inspection on Obstructive Sleep Apnea Severity Detection using a Deep Learning Access

“…The effectiveness of this style for programmed sleep composition was exhibited in our past work. To suit the trip of joint grouping and forecast, we have a tendency to follow the CNN's accepted softmax layer with a perform varied tasks softmax layer and gift the perform varied tasks misfortune work for system making ready.Ye, J. H., Lin, Y., Li, Z. W., Lee, J., Al-Ahmari, A., & Jin, M. [5] present a non-intrusive sleep investigation approach, which may assist U.S. with evaluating the final nature of sleep. Through the flossy derivation, we will get a primer sleep profundity by upgrading the division of sleep organize.…”