Predicting Energy Expenditure During Gradient Walking With a Foot Monitoring Device: Model-Based Approach

Personalized Energy Expenditure Estimation: Visual Sensing Approach With Deep Learning

Background

Calorimetry is both expensive and obtrusive but provides the only way to accurately measure energy expenditure in daily living activities of any specific person, as different people can use different amounts of energy despite performing the same actions in the same manner. Deep learning video analysis techniques have traditionally required a lot of data to train; however, recent advances in few-shot learning, where only a few training examples are necessary, have made developing personalized models without a calorimeter a possibility.

Objective

The primary aim of this study is to determine which activities are most well suited to calibrate a vision-based personalized deep learning calorie estimation system for daily living activities.

Methods

The SPHERE (Sensor Platform for Healthcare in a Residential Environment) Calorie data set is used, which features 10 participants performing 11 daily living activities totaling 4.5 hours of footage. Calorimeter and video data are available for all recordings. A deep learning method is used to regress calorie predictions from video.

Results

Models are personalized with 32 seconds from all 11 actions in the data set, and mean square error (MSE) is taken against a calorimeter ground truth. The best single action for calibration is wipe (1.40 MSE). The best pair of actions are sweep and sit (1.09 MSE). This compares favorably to using a whole 30-minute sequence containing 11 actions to calibrate (1.06 MSE).

Conclusions

A vision-based deep learning energy expenditure estimation system for a wide range of daily living activities can be calibrated to a specific person with footage and calorimeter data from 32 seconds of sweeping and 32 seconds of sitting.

A least action principle for interceptive walking

The principle of least effort has been widely used to explain phenomena related to human behavior ranging from topics in language to those in social systems. It has precedence in the principle of least action from the Lagrangian formulation of classical mechanics. In this study, we present a model for interceptive human walking based on the least action principle. Taking inspiration from Lagrangian mechanics, a Lagrangian is defined as effort minus security, with two different specific mathematical forms. The resulting Euler–Lagrange equations are then solved to obtain the equations of motion. The model is validated using experimental data from a virtual reality crossing simulation with human participants. We thus conclude that the least action principle provides a useful tool in the study of interceptive walking.

Clinometric Gait Analysis Using Smart Insoles in Patients With Hemiplegia After Stroke: Pilot Study

Background

For effective rehabilitation after stroke, it is essential to conduct an objective assessment of the patient’s functional status. Several stroke severity scales have been used for this purpose, but such scales have various limitations.

Objective

Gait analysis using smart insole technology can be applied continuously, objectively, and quantitatively, thereby overcoming the shortcomings of other assessment tools.

Methods

To confirm the reliability of gait analysis using smart insole technology, normal healthy controls wore insoles in their shoes during the Timed Up and Go (TUG) test. The gait parameters were compared with the manually collected data. To determine the gait characteristics of patients with hemiplegia due to stroke, they were asked to wear insoles and take the TUG test; gait parameters were calculated and compared with those of control subjects. To investigate whether the gait analysis accurately reflected the patients’ clinical condition, we analyzed the relationships of 22 gait parameters on 4 stroke severity scales.

Results

The smart insole gait parameter data were similar to those calculated manually. Among the 18 gait parameters tested, 14 were significantly effective at distinguishing patients from healthy controls. The smart insole data revealed that the stance duration on both sides was longer in patients than controls, which has proven difficult to show using other methods. Furthermore, the sound side in patients showed a markedly longer stance duration. Regarding swing duration, that of the sound side was shorter in patients than controls, whereas that of the hemiplegic side was longer. We identified 10 significantly correlated gait parameters on the stroke severity scales. Notably, the difference in stance duration between the sound and hemiplegic sides was significantly correlated with the Fugl-Meyer Assessment (FMA) lower extremity score.

Conclusions

This study confirmed the feasibility and applicability of the smart insole as a device to assess the gait of patients with hemiplegia due to stroke. In addition, we demonstrated that the FMA score was significantly correlated with the smart insole data. Providing an environment where stroke patients can easily measure walking ability helps to maintain chronic functions as well as acute rehabilitation.

Trial Registration

UMIN Clinical Trials Registry UMIN000041646, https://upload.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000047538