Rapid energy expenditure estimation for ankle assisted and inclined loaded walking

Abstract

Background: Estimating energy expenditure with indirect calorimetry requires expensive equipment and several minutes of data collection for each condition of interest. While several methods estimate energy expenditure using correlation to data from wearable sensors, such as heart rate monitors or accelerometers, their accuracy has not been evaluated for activity conditions or subjects not included in the correlation process. The goal of our study was to develop data-driven models to estimate energy expenditure at intervals of approximately one second and demonstrate their ability to predict energetic cost for new conditions and subjects. Model inputs were muscle activity and vertical ground reaction forces, which are measurable by wearable electromyography electrodes and pressure sensing insoles. Methods: We developed models that estimated energy expenditure while walking (1) with ankle exoskeleton assistance and (2) while carrying various loads and walking on inclines. Estimates were made each gait cycle or four second interval. We evaluated the performance of the models for three use cases. The first estimated energy expenditure (in Watts) during walking conditions for subjects with some subject specific training data available. The second estimated all conditions in the dataset for a new subject not included in the training data. The third estimated new conditions for a new subject. Results: The mean absolute percent errors in estimated energy expenditure during assisted walking conditions were 4.4%, 8.0%, and 8.1% for the three use cases, respectively. The average errors in energy expenditure estimation during inclined and loaded walking conditions were 6.1%, 9.7%, and 11.7% for the three use cases. For models not using subject-specific data, we evaluated the ability to order the magnitude of energy expenditure across conditions. The average percentage of correctly ordered conditions was 63% for assisted walking and 87% for incline and loaded walking. Conclusions: We have determined the accuracy of estimating energy expenditure with data-driven models that rely on ground reaction forces and muscle activity for three use cases. For experimental use cases where the accuracy of a data-driven model is sufficient and similar training data is available, standard indirect calorimetry could be replaced. The models, code, and datasets are provided for reproduction and extension of our results.

Sensing leg movement enhances wearable monitoring of energy expenditure

“…Even using comprehensive sensor measurements of leg kinematics and major muscle groups performed similarly to the two selected IMU sensors. The Wearable System was more accurate than wearable data-driven methods using a variety of sensors and subjectspecific training data with errors of 14-27% 16,28,30 , which would have approximately twice the error when evaluating new subjects 35 . When designing wearable devices, rigorous sensor selection may provide counterintuitive results that can significantly improve performance.…”

“…The model was fit using linear regression with ridge regularization with default regularization value of 1. Each stride of data were discretized to a fixed input size by splitting input signals into 30 bins, selected from previous experimentation 35 The Activity-Specific Model relied on ideal activity classification to estimate energy expenditure with separate models for each activity. Inputs to the models consisted of subjects' height, weight, and stride duration.…”

“…Wearable data-driven methods using subject-specific training data have estimated energy expenditure with relatively low errors of 14-27% 16,28,30 . Unfortunately, methods using subject-specific training data have about twice the expected error when estimating energy expenditure for new subjects 35 .…”

The exoskeleton expansion: improving walking and running economy

“…We foresee smart controllers enabling exoskeletons to move beyond conventional fixed assistance parameters, and steering user physiology in-aclosed-loop with the device to maintain optimal exoskeleton assistance across conditions [30,69]. Since measuring metabolic cost throughout everyday life is unrealistic, future exoskeletons may incorporate embedded wearable sensors (e.g., electromyography surface electrodes, pulse oximetry units, and/or low-profile ultrasonography probes) that inform the controller of the user's current physiological state [70,71] and thereby enable continuous optimizing of device assistance [20,72,73] to minimize the user's estimated metabolic cost.…”

Human-in-the-Loop Weight Compensation in Upper Limb Wearable Robots Towards Total Muscles’ Effort Minimization

“…Among all possible biofeedback options, the myoelectric signal is one of the best choices for torque adaptation due to its fast dynamics and monotonic relation with muscle force [20], [21]. At first glance, the myoelectric signal seems to be a noisy signal; however, by using an appropriate signal processing it provides us with more biomechanical information compared to force and motion sensors; it can be used to estimate muscle force [22], fatigue [23], metabolic rate [24], and even discharged timing of motor-neurons [25]. In addition, thanks to recent technology developments, myoelectric signals (e.g., EMG) are considered as accurate, cost-efficient, and small-sized sensors [26] such that the recent commercialized prosthetic hands are benefiting them for real-time motion control [27].…”