The development of an automated assessment system for resistance training movement.

Portable data collection devices and machine learning (ML) have been combined in autonomous movement analysis models for resistance training (RT) movements. However, input features for these models were mostly extracted empirically and subsequent models demonstrated limited interpretability and generalisability to real-world settings. This study aimed to investigate the utility of interpretable and generalisable modelling techniques and several data-driven feature extraction (FE) methods. This was achieved by developing machine learning movement analysis models for the barbell back squat and deadlift using markerless motion capture. 61 participants performed submaximal and maximal repetitions of both RT movements. Movement data was collected using two Azure Kinect cameras. Joint and segment kinematic variables were calculated from the collected depth imaging, and input features were extracted using traditional, manual FE methods and novel data-driven techniques. Classifiers were developed for several predefined technical deviations for both movements. Many of the addressed technical deviations could be classified with good levels of accuracy (≥70%) while the remainder were poor (55%-60%). Additionally, data-driven FE techniques were comparable to previous, traditional FE methods. Interpretable and generalisable modelling techniques can be utilised to good effect for certain classification tasks while data-driven FE techniques did not provide a consistent advantage over traditional FE methods.

Systematic review of automatic assessment systems for resistance-training movement performance: A data science perspective.

The technical performance of resistance-training (RT) movement is commonly monitored through visual assessment and feedback by trained practitioners or by individual self-evaluation. However, both approaches are limited due to their subjectivity, inability to monitor multiple joints simultaneously, and dependency on the assessor's or exerciser's experience and skill. Portable data collection devices and machine learning (ML) have been combined to overcome these limitations by providing objective assessments for RT movement performance. This systematic review evaluates systems developed for providing objective, automatic assessment for RT movements used to improve physical performance and/or rehabilitation in otherwise healthy individuals. Databases searched included Scopus, PubMed and Engineering Village. From 363 papers initially identified, 13 met the inclusion and exclusion criteria. Information extracted from the collated papers included the experimental protocols, data processing, ML model development methodology and movement classification performance. Identified movement assessment systems ranged in classification performance (accuracy of 70%-90% for most classifiers). However, several methodological errors in the development of the ML models were identified, and additional aspects such as model interpretability or generalisability were often neglected. Future ML models should adopt the correct developmental methodology and provide interpretable and generalisable models for application in the RT environment.

Automated robot-assisted assessment for wrist active ranges of motion.

The measurement of wrist active range of motion (ROM) is essential for determining the progress of hand functional recovery, which can provide insight into quantitative improvements and enable effective monitoring during hand rehabilitation. Compared with manual methods, which depend on the experience of the therapist, the proposed robot-assisted assessment technique can measure active ROM of human wrists. The robot with a reconfigurable handle design allows for multiple wrist motions. Experiments were conducted with 11 human subjects to measure ROMs of human wrist flexion/extension and radial/ulnar deviation. Reliability analysis was conducted by calculating the intra-class correlation coefficients (ICC), standard error of measurement (SEM) and SEM%. Results showed high reliability (ICC2,1 ≥ 0.89, SEM ≤ 2.36°, SEM% ≤ 6.81%). Future will focus on adaptive joint self-alignment design between human users and robots to further improve its assessment accuracy.

Validity and Reliability of Upper Limb Functional Assessment Using the Microsoft Kinect V2 Sensor.

OBJECTIVE: To quantify the concurrent accuracy and the test-retest reliability of a Kinect V2-based upper limb functional assessment system. APPROACH: Ten healthy males performed a series of upper limb movements, which were measured concurrently with Kinect V2 and the Vicon motion capture system (gold standard). Each participant attended two testing sessions, seven days apart. Four tasks were performed including hand to contralateral shoulder, hand to mouth, combing hair, and hand to back pocket. Upper limb kinematics were calculated using our developed kinematic model and the UWA model for Kinect V2 and Vicon. The interdevice coefficient of multiple correlation (CMC) and the root mean squared error (RMSE) were used to evaluate the validity of the kinematic waveforms. Mean absolute bias and Pearson's r correlation were used to evaluate the validity of the angles at the points of target achieved (PTA) and the range of motion (ROM). The intersession CMC and RMSE and the intraclass correlation coefficient (ICC) were used to assess the test-retest reliability of Kinect V2. MAIN RESULTS: Both validity and reliability are found to be task-dependent and plane-dependent. Kinect V2 had good accuracy in measuring shoulder and elbow flexion/extension angular waveforms (CMC > 0.87), moderate accuracy of measuring shoulder adduction/abduction angular waveforms (CMC = 0.69-0.82), and poor accuracy of measuring shoulder internal/external angles (CMC < 0.6). We also found high test-retest reliability of Kinect V2 in most of the upper limb angular waveforms (CMC = 0.75-0.99), angles at the PTA (ICC = 0.65-0.91), and the ROM (ICC = 0.68-0.96). SIGNIFICANCE: Kinect V2 has great potential as a low-cost, easy implemented device for assessing upper limb angular waveforms when performing functional tasks. The system is suitable for assessing relative within-person change in upper limb motions over time, such as disease progression or improvement due to intervention.

Reviewing effectiveness of ankle assessment techniques for use in robot-assisted therapy.

This article provides a comprehensive review of studies that investigated ankle assessment techniques to better understand those that can be used in the real-time monitoring of rehabilitation progress for implementation in conjunction with robot-assisted therapy. Seventy-six publications published between January 1980 and August 2013 were selected based on eight databases. They were divided into two main categories (16 qualitative and 60 quantitative studies): 13 goniometer studies, 18 dynamometer studies, and 29 studies about innovative techniques. A total of 465 subjects participated in the 29 quantitative studies of innovative measurement techniques that may potentially be integrated in a real-time monitoring device, of which 19 studies included less than 10 participants. Results show that qualitative ankle assessment methods are not suitable for real-time monitoring in robot-assisted therapy, though they are reliable for certain patients, while the quantitative methods show great potential. The majority of quantitative techniques are reliable in measuring ankle kinematics and kinetics but are usually available only for use in the sagittal plane. Limited studies determine kinematics and kinetics in all three planes (sagittal, transverse, and frontal) where motions of the ankle joint and the subtalar joint actually occur.

An in-vivo lateral ankle ligament strain behavior assessment technique for potential use in robot-assisted therapy.

Ankle sprains are very common, especially in sports activities. Accurate assessment of ankle ligament strain behavior is crucial in understanding ankle function and optimizing ankle rehabilitation programs. This study proposed an in-vivo lateral ankle ligament strain assessment technique for potential use in robot-assisted therapy. It consists of two phases: real-time identification of ankle joint and subtalar joint orientations and simulation of lateral ankle ligament strain behavior. A healthy participant conducted robot-assisted rehabilitation exercises and the results compared to a kinematic model. The model was found to be realistic, leading to the conclusion that this method may be appropriate for determining lateral ankle ligament strain in robot-assisted therapy.

Genetic diversity assessment of sesame core collection in China by phenotype and molecular markers and extraction of a mini-core collection.

BACKGROUND: Sesame (Sesamum indicum L.) is one of the four major oil crops in China. A sesame core collection (CC) was established in China in 2000, but no complete study on its genetic diversity has been carried out at either the phenotypic or molecular level. To provide technical guidance, a theoretical basis for further collection, effective protection, reasonable application, and a complete analysis of sesame genetic resources, a genetic diversity assessment of the sesame CC in China was conducted using phenotypic and molecular data and by extracting a sesame mini-core collection (MC). RESULTS: Results from a genetic diversity assessment of sesame CC in China were significantly inconsistent at the phenotypic and molecular levels. A Mantel test revealed the insignificant correlation between phenotype and molecular marker information (r = 0.0043, t = 0.1320, P = 0.5525). The Shannon-Weaver diversity index (I) and Nei genetic diversity index (h) were higher (I = 0.9537, h = 0.5490) when calculated using phenotypic data from the CC than when using molecular data (I = 0.3467, h = 0.2218). A mini-core collection (MC) containing 184 accessions was extracted based on both phenotypic and molecular data, with a low mean difference percentage (MD, 1.64%), low variance difference percentage (VD, 22.58%), large variable rate of coefficient of variance (VR, 114.86%), and large coincidence rate of range (CR, 95.76%). For molecular data, the diversity indices and the polymorphism information content (PIC) for the MC were significantly higher than for the CC. Compared to an alternative random sampling strategy, the advantages of capturing genetic diversity and validation by extracting a MC using an advanced maximization strategy were proven. CONCLUSIONS: This study provides a comprehensive characterization of the phenotypic and molecular genetic diversities of the sesame CC in China. A MC was extracted using both phenotypic and molecular data. Low MD% and VD%, and large VR% and CR% suggested that the MC provides a good representation of the genetic diversity of the original CC. The MC was more genetically diverse with higher diversity indices and a higher PIC value than the CC. A MC may aid in reasonably and efficiently selecting materials for sesame breeding and for genotypic biological studies, and may also be used as a population for association mapping in sesame.