Wearable Monitoring of Joint Angle and Muscle Activity

Validation of portable in-clinic video-based gait analysis for prosthesis users

Despite the common focus of gait in rehabilitation, there are few tools that allow quantitatively characterizing gait in the clinic. We recently described an algorithm, trained on a large dataset from our clinical gait analysis laboratory, which produces accurate cycle-by-cycle estimates of spatiotemporal gait parameters including step timing and walking velocity. Here, we demonstrate this system generalizes well to clinical care with a validation study on prosthetic users seen in therapy and outpatient clinics. Specifically, estimated walking velocity was similar to annotated 10-m walking velocities, and cadence and foot contact times closely mirrored our wearable sensor measurements. Additionally, we found that a 2D keypoint detector pretrained on largely able-bodied individuals struggles to localize prosthetic joints, particularly for those individuals with more proximal or bilateral amputations, but after training a prosthetic-specific joint detector video-based gait analysis also works on these individuals. Further work is required to validate the other outputs from our algorithm including sagittal plane joint angles and step length. Code for the gait transformer and the trained weights are available at https://github.com/peabody124/GaitTransformer .

Portable, open-source solutions for estimating wrist position during reaching in people with stroke

Arm movement kinematics may provide a more sensitive way to assess neurorehabilitation outcomes than existing metrics. However, measuring arm kinematics in people with stroke can be challenging for traditional optical tracking systems due to non-ideal environments, expense, and difficulty performing required calibration. Here, we present two open-source methods, one using inertial measurement units (IMUs) and another using virtual reality (Vive) sensors, for accurate measurements of wrist position with respect to the shoulder during reaching movements in people with stroke. We assessed the accuracy of each method during a 3D reaching task. We also demonstrated each method's ability to track two metrics derived from kinematics-sweep area and smoothness-in people with chronic stroke. We computed correlation coefficients between the kinematics estimated by each method when appropriate. Compared to a traditional optical tracking system, both methods accurately tracked the wrist during reaching, with mean signed errors of 0.09 ± 1.81 cm and 0.48 ± 1.58 cm for the IMUs and Vive, respectively. Furthermore, both methods' estimated kinematics were highly correlated with each other (p < 0.01). By using relatively inexpensive wearable sensors, these methods may be useful for developing kinematic metrics to evaluate stroke rehabilitation outcomes in both laboratory and clinical environments.

Multi-Arch-Structured All-Carbon Aerogels with Superelasticity and High Fatigue Resistance as Wearable Sensors

Compressible and ultralight all-carbon materials are promising candidates for piezoresistive pressure sensors. Although several fabrication methods have been developed, the required elasticity and fatigue resistance of all-carbon materials are yet to be satisfied as a result of energy loss and structure-derived fatigue failure. Herein, we present a two-stage solvothermal freeze-casting approach to fabricate all-carbon aerogel [modified graphene aerogel (MGA)] with a multi-arched structure, which is enabled by the in-depth solvothermal reduction of graphene oxide and unidirectional ice-crystal growth. MGA exhibits supercompressibility and superelasticity, which can resist an extreme compressive strain of 99% and maintain 93.4% height retention after 100 000 cycles at the strain of 80%. Rebound experiments reveal that MGA can rebound the ball (367 times heavier than the aerogel) in 0.02 s with a very fast recovery speed (∼615 mm s^-1). Even if the mass ratio between the ball and aerogel is increased to 1306, the ball can be rebound in a relatively short time (0.04 s) with a fast recovery speed (∼535 mm s^-1). As a result of its excellent mechanical robustness and electrical conductivity, MGA presents a stable stress-current response (10 000 cycles), tunable linear sensitivity (9.13-7.29 kPa^-1), and low power consumption (4 mW). The MGA-based wearable pressure sensor can monitor human physiological signals, such as pulses, sound vibrations, and muscular movements, demonstrating its potential practicability as a wearable device.