Search

Electrooculogram (EOG) is a well-known physiological metric picked up by placing two or more electrodes around the eyeball. EOG signals are known to be extremely susceptible to motion artifacts. This paper presents a single channel, wireless, wearable flexible EOG monitoring system with motion artifacts sensing and reduction feature. The system uses two non-contact electrode pairs for EOG/motion artifacts detection and motion artifacts reduction. It is implemented on a four-layer flexible polyimide substrate. It is light-weight (only 8.75 gram), battery operated, and uses a microcontroller and a BLE 5.0 transceiver for wireless EOG data transmission, while consuming only 56 mW of power. The system metrics such as gain around 37 dB, bandwidth from 1 Hz to 40 Hz, and noise are evaluated. The system is tested for different electrode configurations and it is demonstrated that horizontally parallel electrode pairs achieve an acceptable motion artifact reduction at the output, while preserving perfect EOG features (such as eye-blinking). The average sensitivity for horizontally parallel non-contact electrodes is found out to be more than 50 times with respect to commercial gold electrodes, whereas the average response time of the sensor is around 380 mS. The flexible EOG system is comfortable to wear and the use of non-contact electrode eliminates the need of skin preparation. Therefore, the system can be easily integrated with eye-masks and headbands, thus making it an excellent prototype for many smart applications. [ABSTRACT FROM AUTHOR]

Robustness of estimating cardiorespiratory parameters from photoplethysmography (PPG) signal is highly dependent on the quality of the signal, which is heavily affected by motion artifacts. To increase the estimation accuracy of cardiorespiratory parameters, this article describes a novel fusion method to efficiently and effectively reduce the motion artifacts from the acquired PPG signal. The proposed fusion technique requires simultaneously acquiring data from a PPG sensor and accelerometer. To filter out the frequencies associated with motion, the method uses stopband filters with a central rejection frequency and bandwidth determined by the output signal of the accelerometer. Under such conditions, the proposed method to remove the motion artifacts does not depend on the quality of the reference signal and has almost no impact on the nature of PPG signals (i.e., amplitude, baseline, and periodicity). The effectiveness of the proposed method in the suppression of in-band and out-of-band frequencies of motion is numerically and experimentally evaluated. It is shown that the filtered PPG signal has sufficient information to estimate different cardiac parameters such as heart- (HR), respiration rate (RR), and blood oxygen saturation (SpO2). The motion artifact-free PPG signal obtained using our proposed method can estimate HR, RR, and SpO2 with an accuracy of above 95%. This level of accuracy confirms the usefulness of the proposed fusion method for accuracy improvement of cardiorespiratory parameters monitored by the filtered PPG signal. [ABSTRACT FROM AUTHOR]

The ability to assess the respiratory rate reliably is vital in the application of remote health monitoring. A newly developed pulmonary monitoring system based on diaphragm wall motion tracking is evaluated under human physical activity. Diaphragm excursions are tracked by a designed ultrasound sensory system utilizing three ultrasound PZT5 piezo transducers. We assess the accuracy and reliability of this system in monitoring the diaphragmatic function and its contribution to the respiratory workload. We also examine inertial and pulse-oximetry sensors as two alternative methods. Measurements are compared to a spirometer as the gold standard. The tests are designed in this study to investigate the performance of employed sensors in both the stationary and non-stationary human body situations. We conclude that by the direct tracking of the diaphragm motion, the proposed ultrasound system is fundamentally robust to motion artifacts. Outstanding results were obtained for the proposed system from six subjects in six different test models with an average sensitivity, specificity and false alarm of 89.7%, 93.1%, and 6.3%, respectively. [ABSTRACT FROM AUTHOR]