U2PNet: An Unsupervised Underwater Image-Restoration Network Using Polarization.

This article presents U2PNet, a novel unsupervised underwater image restoration network using polarization for improving signal-to-noise ratio and image quality in underwater imaging environments. Traditional methods for underwater image restoration using polarization require specific cues or pairs of underwater polarization datasets, which limit their practical applications. Our proposed method requires only one mosaicked polarized image of the scene and does not require datasets for pretraining or specific cues. We design two subnetworks (T-net and B ∞ -net) to accurately estimate the transmission map and background light, and unique nonreference loss functions to ensure effective restoration. Our experiments are based on an indoor polarization simulated dataset and a real polarization image dataset constructed from our underwater robotic platform equipped with polarization cameras. Experiment results demonstrate that our proposed method achieves state-of-the-art performance on both simulated and real underwater polarization images. The code and datasets will be available at https://github.com/polwork/U-2Pnet.

Shaping high-performance wearable robots for human motor and sensory reconstruction and enhancement.

Most wearable robots such as exoskeletons and prostheses can operate with dexterity, while wearers do not perceive them as part of their bodies. In this perspective, we contend that integrating environmental, physiological, and physical information through multi-modal fusion, incorporating human-in-the-loop control, utilizing neuromuscular interface, employing flexible electronics, and acquiring and processing human-robot information with biomechatronic chips, should all be leveraged towards building the next generation of wearable robots. These technologies could improve the embodiment of wearable robots. With optimizations in mechanical structure and clinical training, the next generation of wearable robots should better facilitate human motor and sensory reconstruction and enhancement.

Multi-day monitoring of foot progression angles during unsupervised, real-world walking in people with and without knee osteoarthritis.

BACKGROUND: Foot progression angle is a biomechanical target in gait modification interventions for knee osteoarthritis. To date, it has only been evaluated within laboratory settings. METHODS: Adults with symptomatic knee osteoarthritis (n = 30) and healthy adults (n = 15) completed two conditions: 1) treadmill walking in the laboratory (5-min), and 2) real-world walking outside of the laboratory (1-week). Foot progression angle was estimated via shoe-embedded inertial sensing. We calculated the foot progression angle magnitude (median) and variability (interquartile range, coefficient of variation), and used mixed models to compare outcomes between the conditions, participant groups, and disease severities. Reliability was quantified by the intraclass correlation coefficient, standardized error of the measurement, and the minimum detectable change. FINDINGS: Foot progression angle magnitude did not differ between groups or conditions but variability significantly higher in real-world walking (P < 0.001). Structural and symptomatic severity were unrelated to FPA in either walking condition, except for real-world coefficient of variation which was higher for moderate-severe structural osteoarthritis compared to the treadmill for those with mild structural severity (P < 0.034). All real-world outcomes showed excellent reliability including intraclass correlation coefficients above 0.95. The participants recorded a mean (standard deviation) of 298 (33) and 10,447 (5232) steps in the laboratory and real-world walking conditions, respectively. INTERPRETATION: This study provides the first characterization of foot progression angles during real-world walking in people with and without symptomatic knee osteoarthritis. These results indicate that foot progression angles can be feasibly and reliably measured in unsupervised real-world walking conditions.

Human-in-the-Loop Adaptive Control of a Soft Exo-Suit With Actuator Dynamics and Ankle Impedance Adaptation.

Soft exo-suit could facilitate walking assistance activities (such as level walking, upslope, and downslope) for unimpaired individuals. In this article, a novel human-in-the-loop adaptive control scheme is presented for a soft exo-suit, which provides ankle plantarflexion assistance with unknown human-exosuit dynamic model parameters. First, the human-exosuit coupled dynamic model is formulated to express the mathematical relationship between the exo-suit actuation system and the human ankle joint. Then, a gait detection approach, including plantarflexion assistance timing and planning, is proposed. Inspired by the control strategy that is used by the human central nervous system (CNS) to handle interaction tasks, a human-in-the-loop adaptive controller is proposed to adapt the unknown exo-suit actuator dynamics and human ankle impedance. The proposed controller can emulate human CNS behaviors which adapt feedforward force and environment impedance in interaction tasks. The resulting adaptation of actuator dynamics and ankle impedance is demonstrated with five unimpaired subjects and implemented on a developed soft exo-suit. The human-like adaptivity is performed by the exo-suit in several human walking speeds and illustrates the promising potential of the novel controller.

Wearable Real-Time Haptic Biofeedback Foot Progression Angle Gait Modification to Assess Short-Term Retention and Cognitive Demand.

Foot progression angle gait (FPA) modification is an important part of rehabilitation for a variety of neuromuscular and musculoskeletal diseases. While wearable haptic biofeedback could enable FPA gait modification for more widespread use than traditional tethered, laboratory-based approaches, retention, and cognitive demand in FPA gait modification via wearable haptic biofeedback are currently unknown and may be important to real-life implementation. Thus, the purpose of this study was to assess the feasibility of wearable haptic biofeedback to assess short-term retention and cognitive demand during FPA gait modification. Ten healthy participants performed toe-in (target 10 degrees change in internal rotation) and toe-out (target 10 degrees change in external rotation) haptic gait training trials followed by short-term retention trials, and cognitive multitasking trials. Results showed that participants were able to initially respond to the wearable haptic feedback to modify their FPA to adopt the new toe-in (9.7 ± 0.8 degree change in internal rotation) and toe-out (8.9 ± 1.0 degree change in external rotation) gait patterns. Participants retained the modified gait pattern on average within 3.9 ± 3.6 deg of the final haptic gait training FPA values. Furthermore, cognitive multitasking did not influence short-term retention in that there were no differences in gait performance during retention trials with or without cognitive multitasking. These results demonstrate that wearable haptic biofeedback can be used to assess short-term retention and cognitive demand during FPA gait modification without the need for traditional, tethered systems.

IMU-based knee flexion, abduction and internal rotation estimation during drop landing and cutting tasks.

Anterior cruciate ligament (ACL) injury is a common and severe knee injury in sports. Knee flexion, abduction and internal rotation angles are considered crucial biomechanical indicators of the ACL injury risk but currently are computed in a laboratory with an optical motion capture. This paper introduces an inertial measurement unit (IMU) based algorithm for knee flexion, abduction and internal rotation estimation during ACL injury risk assessment tests, including drop landing and cutting tasks. This algorithm includes a special two-step complementary-based orientation filter and a special single-pose sensor-to-segment calibration procedure. Fourteen healthy subjects performed double-leg, single-leg drop landing and cutting tasks. Each subject wore four IMUs and reflective marker clusters on their thighs and shanks. For the presented knee angles algorithm with an empirical initial segment orientation, the root mean square errors (RMSEs) of the estimated continuous knee flexion, abduction and internal rotation cross all the movement tasks were 1.07°, 2.87° and 2.64°, and RMSEs of the peak knee flexion and peak knee abduction errors were 1.22° and 3.82°. The knee angles algorithm was capable of estimating knee abduction and internal rotation angles during drop landing and cutting tasks, and knee flexion estimation was substantially more accurate than previously reported approaches. Additionally, we found that for the presented algorithm, the accuracy of initial segment orientation was a critical factor for knee abduction and internal rotation estimations. The presented IMU-based knee angles algorithm could serve as a foundation to enable in-field biomechanical ACL injury risk assessment.

Self-selected running gait modifications reduce acute impact loading, awkwardness, and effort.

Impact loading has been associated with running-related injuries, and gait retraining has been suggested as a means of reducing impact loading and lowering the risk of injury. However, gait retraining can lead to increased perceived awkwardness and effort. The influence of specifically trained and self-selected running gait modifications on acute impact loading, perceived awkwardness and effort is currently unclear. Sixteen habitual rearfoot/midfoot runners performed forefoot strike pattern, increased step rate, anterior trunk lean and self-selected running gait modifications on an instrumented treadmill based on real-time biofeedback. Impact loading, perceived awkwardness and effort scores were compared among the four gait retraining conditions. Self-selected gait modification reduced vertical average loading rate (VALR) by 25.3%, vertical instantaneous loading rate (VILR) by 27.0%, vertical impact peak (VIP) by 16.8% as compared with baseline. Forefoot strike pattern reduced VALR, VILR and peak tibial acceleration. Increased step rate reduced VALR. Anterior trunk lean did not reduce any impact loading. Self-selected gait modification was perceived as less awkward and require less effort than the specifically trained gait modification (p < 0.05). These findings suggest that self-selected gait modification could be a more natural and less effortful strategy than specifically trained gait modification to reduce acute impact loading, while the clinical significance remains unknown.

Magnetometer-Free, IMU-Based Foot Progression Angle Estimation for Real-Life Walking Conditions.

Foot progression angle (FPA) is vital in many disease assessment and rehabilitation applications, however previous magneto-IMU-based FPA estimation algorithms can be prone to magnetic distortion and inaccuracies after walking starts and turns. This paper presents a foot-worn IMU-based FPA estimation algorithm comprised of three key components: orientation estimation, acceleration transformation, and FPA estimation via peak foot deceleration. Twelve healthy subjects performed two walking experiments to evaluation IMU algorithm performance. The first experiment aimed to validate the proposed algorithm in continuous straight walking tasks across seven FPA gait patterns (large toe-in, medium toe-in, small toe-in, normal, small toe-out, medium toe-out, and large toe-out). The second experiment was performed to evaluate the proposed FPA algorithm for steps after walking starts and turns. Results showed that FPA estimations from the IMU-based algorithm closely followed marker-based system measurements with an overall mean absolute error of 3.1±1.3 deg, and the estimation results were valid for all steps immediately after walking starts and turns. This work could enable FPA assessment in environments where magnetic distortion is present due to ferrous metal structures and electrical equipment, or in real-life walking conditions when walking starts, stops, and turns commonly occur.

Modeling and Prediction of Wearable Energy Harvesting Sliding Shoes for Metabolic Cost and Energy Rate Outside of the Lab.

The recent explosion of wearable electronics has led to widespread interest in harvesting human movement energy, particularly during walking, for clinical and health applications. However, the amount of energy available to harvest and the required metabolic rate for wearable energy harvesting varies across subjects. In this paper, we utilize custom energy harvesting sliding shoes to develop and evaluate multivariate linear regression models to predict metabolic rate and energy harvesting rate during overground walking outside of the lab. Subjects performed 200 m self-selected normal and fast walking trials on flat ground with custom sliding shoes. Metabolic rate was measured with a portable breathing analysis system and energy harvesting rate was measured directly from the generator on the custom sliding shoes. Model performance was determined by comparing the difference between actual and predicted metabolic and energy harvesting rates. Overall, predictive modeling closely matched the actual values, and there was no statistical difference between actual and predicted average metabolic rate or between actual and predicted average energy harvesting rate. Energy harvesting sliding shoes could potentially be used for a variety of wearable devices to reduce onboard energy storage, and these findings could serve to inform expected energy harvesting rates and associated required metabolic cost for a diverse array of medical and health applications.

Portable, automated foot progression angle gait modification via a proof-of-concept haptic feedback-sensorized shoe.

Modifying the foot progression angle (FPA) is a non-pharmacological, non-surgical treatment option for knee osteoarthritis, however current widespread adoption has been limited by the requirement of laboratory-based motion capture systems. We present the first customized haptic feedback-sensorized shoe for estimating and modifying FPA during walking gait, which includes an electronic inertial and magnetometer module in the sole for estimating FPA, and two vibration motors attached to the medial and lateral shoe lining for providing vibrotactile feedback. Feasibility testing was performed by comparing FPA performance while wearing the haptic feedback-sensorized shoe with the training targets. Participants performed five walking trials with five randomly-presented FPA targets (10° toe-in, 0°, 10° toe-out, 20° toe-out, and 30° toe-out) of 2 min each on a treadmill. Overall average FPA performance error across all conditions was 0.2 ± 4.1°, and the overall mean absolute FPA performance error across all conditions was 3.1 ± 2.6°. Reducing the size of the no-feedback window resulted in less performance error during walking. This study demonstrates that a novel haptic feedback-sensorized shoe can be used to effectively train FPA modifications. The haptic feedback-sensorized shoe could potentially be used for FPA gait modification outside of specialized camera-based motion capture laboratories as a conservative treatment for knee osteoarthritis or other related clinical applications requiring FPA assessment and modification in daily life.

"Controlled Slip" Energy Harvesting While Walking.

Rapid development in wearable electronics and systems continues to impose challenges on portable energy storage sustained over time, and thus human energy harvesting is a potentially attractive means of sustainable, long-term energy. We introduce a novel 'controlled slip' energy harvesting approach for capturing energy during human locomotion. While slip is normally considered undesirable, controlled slip holds potential to enable a significant amount of energy harvesting during each step of human gait. Custom-designed 'controlled slip' energy harvesting shoes were fabricated by mounting a sliding plate and a generator with a one-way bearing to the sole of standard walking shoes, which induces controlled forward slip during early stance while energy is harvested. Fourteen healthy subjects performed treadmill walking trials with the 'controlled slip' energy harvesting shoes which generated average electrical power of 1.15-1.44 W at walking speeds of 2.9-4.3 km/h. Interestingly, without prompting, subjects chose to walk with the 'controlled slip' energy harvesting shoes in either one of two distinct ways: landing with the heel first (heel strikers) or landing with the toe first (toe strikers). While heel strikers and toe strikers exhibited similar electrical power output and hip flexion angle at initial foot contact, heel strikers had higher peak ankle power and lower knee flexion angle at initial foot contact than toe strikers. 'Controlled slip' energy harvesting could potentially generate electrical power for a broad spectrum of wearable devices.

Validity and reliability of a shoe-embedded sensor module for measuring foot progression angle during over-ground walking.

Wearable systems are becoming increasingly popular for gait assessment outside of laboratory settings. A single shoe-embedded sensor module can measure the foot progression angle (FPA) during walking. The FPA has important clinical utility, particularly in populations with knee osteoarthritis, as it is a target for biomechanical treatments. However, the validity and the day-to-day reliability of FPA measurement using wearable systems during over-ground walking has yet to be established. Two gait analysis sessions on 20 healthy adults were conducted. During both sessions, participants performed natural over-ground walking in a motion capture laboratory and on a 100 m linear section of outdoor athletics track. FPA was measured in the laboratory via marker trajectory data, while the sensor module measured FPA during the outdoor track walking. Validity was examined by comparing the laboratory- and sensor-measured average FPA. Day-to-day reliability was examined by comparing the sensor-measured FPA between the first and second gait analysis sessions. Average absolute error between motion capture and sensor measured FPA were 1.7° and 2.1° at session 1 and 2, respectively. A Bland and Altman plot indicated no systematic bias, with 95% limit of agreement widths of 4.2° - 5.1°. Intraclass correlation coefficient (ICC2k) analysis resulted in good to excellent validity (ICC = 0.89 - 0.91) and reliability (ICC = 0.95). Overall, the shoe-embedded sensor module is a valid and reliable method of measuring FPA during over-ground walking without the need for laboratory equipment.

Foot strike pattern, step rate, and trunk posture combined gait modifications to reduce impact loading during running.

Elevated impact loading can be detrimental to runners as it has been linked to the increased risk of tibial stress fracture and plantar fasciitis. The objective of this study was to investigate the combined effects of foot strike pattern, step rate, and anterior trunk lean gait modifications on impact loading in runners. Nineteen healthy runners performed 12 separate gait modification trials involving: three foot strike patterns (rearfoot, midfoot, and forefoot strike), two step rates (natural and 10% increased), and two anterior trunk lean postures (natural and 10-degree increased flexion). Overall, forefoot strike combined with increased step rate led to the lowest impact loading rates, and rearfoot strike combined with anterior trunk lean led to the highest impact loading rates. In addition, there were interaction effects between foot strike pattern and step rate on awkwardness and effort, such that it was both more natural and easier to transition to a combined gait modification involving forefoot strike and increased step rate than to an isolated gait modification involving either forefoot strike or increased step rate. These findings could help to inform gait modifications for runners to reduce impact loading and associated injury risks.

Validation of a smart shoe for estimating foot progression angle during walking gait.

The foot progression angle is an important measurement related to knee loading, pain, and function for individuals with knee osteoarthritis, however current measurement methods require camera-based motion capture or floor-embedded force plates confining foot progression angle assessment to facilities with specialized equipment. This paper presents the validation of a customized smart shoe for estimating foot progression angle during walking. The smart shoe is composed of an electronic module with inertial and magnetometer sensing inserted into the sole of a standard walking shoe. The smart shoe charges wirelessly, and up to 160h of continuous data (sampled at 100Hz) can be stored locally on the shoe. For validation testing, fourteen healthy subjects were recruited and performed treadmill walking trials with small, medium, and large toe-in (internal foot rotation), small, medium, and large toe-out (external foot rotation) and normal foot progression angle at self-selected walking speeds. Foot progression angle calculations from the smart shoe were compared with measurements from a standard motion capture system. In general, foot progression angle values from the smart shoe closely followed motion capture values for all walking conditions with an overall average error of 0.1±1.9deg and an overall average absolute error of 1.7±1.0deg. There were no significant differences in foot progression angle accuracy across the seven different walking gait patterns. The presented smart shoe could potentially be used for knee osteoarthritis or other clinical applications requiring foot progression angle assessment in community settings or in clinics without specialized motion capture equipment.