Estimating load positions of wearable devices based on difference in pulse wave arrival time

Yoshida, Kazuki; Murao, Kazuya

doi:10.1145/3341163.3347743

Cited by 6 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From the collected data, the attachment site was estimated using the C4.5 classifier. In addition, Yoshida et al [34] proposed a method to estimate the body part where a wearable device is attached without requiring the wearer to perform a specific action; instead, they used electrocardiography (ECG) and pulse data, which are biometric information that can be acquired by the wearable device.…”

Section: A Sensing With Wearable Devicesmentioning

confidence: 99%

Pulse Wave Generation Method for PPG by Using Display

2023

Self Cite

View full text Add to dashboard Cite

The extensive research on wearable devices has led to devices with various shapes and mounting locations. Wearable devices are often used to record the user's biometric information, and methods have been proposed to detect physical abnormalities from the acquired data. Among various kinds of biometric data, pulse data has been used in methods such as heart rate monitoring and emotion recognition. The most common type of pulse sensor uses photoplethysmography (PPG), which irradiates a green LED on the skin and measures pulse data from changes in the light reflected from blood vessels. PPG sensors have been implemented in commercially available wearable devices such as smartwatches. However, a PPG sensor requires blood flow for data acquisition, and when a smartwatch is worn on an artificial body part such as a prosthetic hand or a robotic arm, data cannot be acquired because there is no blood flow. In this study, we propose a method that enables a PPG sensor to measure arbitrary pulse data by using a display. If this method is successful, it will enable pulse data measured at the junction of a living limb and an artificial limb to be input to the display; then, a smartwatch attached to the artificial limb will read the same pulse data. In this paper, we focus on the heart rate and report the results of an experiment in which a target heart rate was input and the display was controlled accordingly to determine whether the target heart rate could be obtained by a smartwatch. We implemented a display drawing program and conducted the evaluation with five kinds of smartwatches and four kinds of displays. The results showed that the error between the target heart rate and the heart rate acquired by the smartwatch was within 3 beats per minute (bpm) in many cases when the target heart rate was set to 60-100 bpm. When the target heart rate was set to 40-55 and 105-200 bpm, the heart rate could also be input to the smartwatch with a small error under certain conditions. Moreover, when generated PPG data was imported into heart rate variability (HRV) analysis software, it was recognized as a pulse wave in the same way as real PPG data obtained from a person. We compared the heart rate, RR interval, and SDNN calculated from the real and generated PPG data, and we confirmed that the proposed method could generate stable PPG data. On the other hand, when the waveforms were compared, the generated PPG waveform differed greatly from the real PPG waveform, which indicated that the software could calculate the heart rate, RR interval, SDNN, and LF/HF ratio regardless of the waveform. This result suggests that calculation of these parameters without verifying the waveform would be vulnerable to an attack with fake PPG data.

show abstract

Section: A Sensing With Wearable Devicesmentioning

confidence: 99%

Pulse Wave Generation Method for PPG by Using Display

2023

Self Cite

View full text Add to dashboard Cite

show abstract

“…When benchmarked on a held-out test set, SleepPPG-Net obtained a median Cohen’s Kappa score of 0.75 against 0.69 for the best SOTA approach. Yoshida et al [ 25 ] proposed a method for estimating the load position of wearable devices using pulse waves and ECG. They estimate the arrival time of the pulse wave at the wearable device mounting position by comparing the heartbeat obtained by the ECG sensor with the pulse wave obtained by the pulse wave sensor at load position and estimate the load position of wearable devices by calculating the KL divergence between the distribution of the estimated time and the distribution of the training data collected at each site in advance.…”

Section: Related Workmentioning

confidence: 99%

PPG2EMG: Estimating Upper-Arm Muscle Activities and EMG from Wrist PPG Values

Okamoto

Murao

2023

Sensors

Self Cite

View full text Add to dashboard Cite

The electromyogram (EMG) is a waveform representation of the action potential generated by muscle cells using electrodes. EMG acquired using surface electrodes is called surface EMG (sEMG), and it is the acquisition of muscle action potentials transmitted by volume conduction from the skin. Surface electrodes require disposable conductive gel or adhesive tape to be attached to the skin, which is costly to run, and the tape is hard on the skin when it is removed. Muscle activity can be evaluated by acquiring muscle potentials and analyzing quantitative, temporal, and frequency factors. It is also possible to evaluate muscle fatigue because the frequency of the EMG becomes lower as the muscle becomes fatigued. Research on human activity recognition from EMG signals has been actively conducted and applied to systems that support arm and hand functions. This paper proposes a method for recognizing the muscle activity state of the arm using pulse wave data (PPG: Photoplethysmography) and a method for estimating EMG using pulse wave data. This paper assumes that the PPG sensor is worn on the user’s wrist to measure the heart rate. The user also attaches an elastic band to the upper arm, and when the user exerts a force on the arm, the muscles of the upper arm contract. The arteries are then constricted, and the pulse wave measured at the wrist becomes weak. From the change in the pulse wave, the muscle activity of the arm can be recognized and the number of action potentials of the muscle can be estimated. From the evaluation experiment with five subjects, three types of muscle activity were recognized with 80+%, and EMG was estimated with approximately 20% error rate.

show abstract

“…Two papers focused their sensing on and around the heart. Best Paper winners Yoshida et al introduced a unique new way to resolve the challenge of not knowing exactly where on the body a sensor may have been placed [12]. Using the fact that electricity moves faster than blood flow, they estimate location by measuring the difference in time between an electrocardiogram-sensed heartbeat and the corresponding pulse-wave detected at the sensor location.…”

Section: Wearable Sensingmentioning

confidence: 99%