The objective is to develop a cuffless method that accurately estimates blood pressure (BP) during activities of daily living. User-specific nonlinear autoregressive models with exogenous inputs (NARX) are implemented using artificial neural networks to estimate the BP waveforms from electrocardiography and photoplethysmography signals. To broaden the range of BP in the training data, subjects followed a short procedure consisting of sitting, standing, walking, Valsalva maneuvers, and static handgrip exercises. The procedure was performed before and after a six-hour testing phase wherein five participants went about their normal daily living activities. Data were further collected at a four-month time point for two participants and again at six months for one of the two. The performance of three different NARX models was compared with three pulse arrival time (PAT) models. The NARX models demonstrate superior accuracy and correlation with “ground truth” systolic and diastolic BP measures compared to the PAT models and a clear advantage in estimating the large range of BP. Preliminary results show that the NARX models can accurately estimate BP even months apart from the training. Preliminary testing suggests that it is robust against variabilities due to sensor placement. This establishes a method for cuffless BP estimation during activities of daily living that can be used for continuous monitoring and acute hypotension and hypertension detection.
19 September 2022 in IEEE Journal of Biomedical and Health Informatics
Objective: To develop and evaluate an accurate method for cuffless blood pressure (BP) estimation during moderate- and heavy-intensity exercise. Methods: Twelve participants performed three cycling exercises: a ramp-incremental exercise to exhaustion, and moderate and heavy pseudorandom binary sequence exercises on an electronically braked cycle ergometer over the course of 21 minutes. Subject-specific and population-based nonlinear autoregressive models with exogenous inputs (NARX) were compared with feedforward artificial neural network (ANN) models and pulse arrival time (PAT) models. Results: Population-based NARX models, that is, trained on the three cycling exercise conditions for 11 participants and tested on the remaining participant (applying leave-one-subject-out cross-validation), performed better than the other models and showed good capability for estimating large changes in mean arterial pressure (MAP). The models were unable to track consistent decreases in BP during prolonged exercise caused by reduction in peripheral vascular resistance, since this information is apparently not encoded in the employed proxy physiological signals (electrocardiography and forehead PPG) used for BP estimation. Nevertheless, the population-based NARX model had an error standard deviation of 11.0 mmHg during the entire exercise window, which improved to 9.0 mmHg when the model was periodically calibrated every 7 minutes. Conclusion: Population-based NARX models can estimate BP during moderate- and heavy-intensity exercise but need periodic calibration to account for the change in vascular resistance during exertion. Significance: MAP can be continuously tracked during exercise using only wearable sensors, making monitoring exercise physiology more convenient and accessible.
Objective: Oscillometric finger pressing is a smartphone-based blood pressure (BP) monitoring method. Finger photoplethysmography (PPG) oscillations and pressure are measured during a steady increase in finger pressure, and an algorithm computes systolic BP (SP) and diastolic BP (DP) from the measurements. The objective was to assess the impact of finger artery viscoelasticity on the BP computation. Methods: Nonlinear viscoelastic models relating transmural pressure (finger BP – applied pressure) to PPG oscillations during finger pressing were developed. The output of each model to a measured transmural pressure input was fitted to measured PPG oscillations from 15 participants. A parametric sensitivity analysis was performed via model simulations to elucidate the viscoelastic effect on the derivative-based BP computation algorithm. Results: A Wiener viscoelastic model comprising a first-order transfer function followed by a static sigmoidal function fitted the measured PPG oscillations better than an elastic model containing only the static function (median (IQR) error of 30.5% (25.6%–34.0%) vs 50.9% (46.7%–53.7%); p<0.01). In Wiener model simulations, the derivative algorithm underestimated SP, especially with high pulse pressure and low transfer function cutoff frequency (i.e., greater viscoelasticity). The mean of the normalized PPG waveform at the maximum oscillation beat was found to correlate with the cutoff frequency ( r = −0.8) and could thus possibly be used to compensate for viscoelasticity. Conclusion: Finger artery viscoelasticity negatively impacts oscillometric BP computation algorithms but can potentially be compensated for using available measurements. Significance: These findings may help in converting smartphones into truly cuffless BP monitors for improving hypertension awareness and control.
Objectives: Grip force during hand tool operation is the primary contributor to tendon strain and related wrist injuries, whereas push force is a contributor to shoulder injuries. However, both cannot be directly measured using a single measurement instrument. The objective of this research was to develop and test an algorithm to isolate the grip and push force distributions from in-situ hand-handle pressure measurements and to quantify their distributions among industrial workers using an electric nutrunner. Methods: Experienced automobile assembly line workers used an industrial nutrunner to tighten fasteners at various locations and postures. The pressure applied by the hand on the tool handle was measured dynamically using pressure sensors mounted on the handle. An algorithm was developed to compute the push force applied to the handle of an electric pistol-grip nutrunner based on recorded pressure measurements. An optimization problem was solved to find the contribution of each measured pressure to the actual pushing force of the tool. Finally, the grip force was determined from the difference between the measured pressure and the calculated pushing pressure. Results: The grip force and push force were successfully isolated and there was no correlation between the two forces. The computed grip force increased from low to high fastener locations, whereas the push force significantly increased during overhead fastening. A significant difference across the participants’ computed grip forces was observed. The grip force distribution showed that its contribution to total hand force was larger than other definitions in the literature. Conclusions: The developed algorithm can aid in better understanding the risk of injury associated with different tasks through the notion of grip and push force distribution. This was shown to be important as even workers with considerable power tool experience applied significantly more grip and push force than other participants, all of whom successfully completed each task. Moreover, the fact that both forces were uncorrelated shows the need for extracting them independently.
5 July 2024 in IEEE Journal of Biomedical and Health Informatics
Intermittent pneumatic compression (IPC) systems apply external pressure to the lower limbs and enhance peripheral blood flow. We previously introduced a cardiac-gated compression system that enhanced arterial blood velocity (BV) in the lower limb compared to fixed compression timing (CT) for seated and standing sub7 jects. However, these pilot studies found that the CT that maximized BV was not constant across individuals and could change over time. Current CT modelling methods for IPC are limited to predictions for a single day and one heartbeat ahead. However, IPC therapy for may span weeks or longer, the BV response to compression can vary with physiological state, and the best CT for eliciting the desired physiological outcome may change, even for the same individual. We propose that a deep reinforcement learning (DRL) algorithm can learn and adaptively modify CT to achieve a selected outcome using IPC. Herein, we target maximizing lower limb arterial BV as the desired out19 come and build participant-specific simulated lower limb environments for 6 participants. We show that DRL can adaptively learn the CT for IPC that maximized arterial BV. Compared to previous work, the DRL agent achieves 98% ± 2 of the resultant blood flow and is faster at maximizing BV; the DRL agent can learn an “optimal” policy in 15 minutes ± 2 on average and can adapt on the fly. Given a desired objective, we posit that the proposed DRL agent can be implemented in IPC systems to rapidly learn the (potentially time-varying) “optimal” CT with a human-in-the-loop.