The use of artificial intelligence in the analysis of sports performance: a review of applications in human gait analysis and future directions for sports biomechanics

Systematic review of automatic post-stroke gait classification systems

BACKGROUND

Gait classification is a clinically helpful task performed after a stroke in order to guide rehabilitation therapy. Gait disorders are commonly identified using observational gait analysis in clinical settings, but this approach is limited due to low reliability and accuracy. Data-driven gait classification can quantify gait deviations and categorise gait patterns automatically possibly improving reliability and accuracy; however, the development and clinical utility of current data driven systems has not been reviewed previously.

RESEARCH QUESTION

The purpose of this systematic review is to evaluate the literature surrounding the methodology used to develop automatic gait classification systems, and their potential effectiveness in the clinical management of stroke-affected gait.

METHOD

The database search included PubMed, IEEE Xplore, and Scopus. Twenty-one studies were identified through inclusion and exclusion criteria from 407 available studies published between 2015 and 2022. Development methodology, classification performance, and clinical utility information were extracted for review.

RESULTS AND SIGNIFICANCE

Most of gait classification systems reported a classification accuracy between 80%-100%. However, collated studies presented methodological errors in machine learning (ML) model development. Further, many studies neglected model components such as clinical utility (e.g., predictions don't assist clinicians or therapists in making decisions, interpretability, and generalisability). We provided recommendations to guide development of future post-stroke automatic gait classification systems to better assist clinicians and therapists. Future automatic gait classification systems should emphasise the clinical significance and adopt a standardised development methodology of ML model.

Artificial Intelligence Approach in Biomechanics of Gait and Sport: A Systematic Literature Review

Background

Artificial neural network helps humans in a wide range of activities, such as sports.

Objective

This paper aims to investigate the effect of artificial intelligence on decision-making related to human gait and sports biomechanics, using computer-based software, and to investigate the impact of artificial intelligence on individuals' biomechanics during gait and sports performance.

Material and Methods

This review was conducted in compliance with the PRISMA guidelines. Abstracts and citations were identified through a search based on Science Direct, Google Scholar, PubMed, Elsevier, Springer Link, Web of Science, and Scopus search engines from 1995 up to 2023 to obtain relevant literature about the impact of artificial intelligence on biomechanics. A total of 1000 articles were found related to biomechanical characteristics of gait and sport and 26 articles were directly pertinent to the subject.

Results

The extent of the application of artificial intelligence in sports biomechanics in various fields. In addition, various variables in the fields of kinematics, kinetics, and the field of time can be investigated based on artificial intelligence. Conventional computational techniques are limited by the inability to process data in its raw form. Artificial Intelligence (AI) and Machine Learning (ML) techniques can handle complex and high-dimensional data.

Conclusion

The utilization of specialized systems and neural networks in gait analysis has shown great potential in sports performance analysis. Integrating AI into this field would be a significant advancement in sport biomechanics. Coaches and athletes can develop more precise training regimens with specialized performance prediction models.

Using Artificial Intelligence-Enhanced Sensing and Wearable Technology in Sports Medicine and Performance Optimisation

Wearable technologies are small electronic and mobile devices with wireless communication capabilities that can be worn on the body as a part of devices, accessories or clothes. Sensors incorporated within wearable devices enable the collection of a broad spectrum of data that can be processed and analysed by artificial intelligence (AI) systems. In this narrative review, we performed a literature search of the MEDLINE, Embase and Scopus databases. We included any original studies that used sensors to collect data for a sporting event and subsequently used an AI-based system to process the data with diagnostic, treatment or monitoring intents. The included studies show the use of AI in various sports including basketball, baseball and motor racing to improve athletic performance. We classified the studies according to the stage of an event, including pre-event training to guide performance and predict the possibility of injuries; during events to optimise performance and inform strategies; and in diagnosing injuries after an event. Based on the included studies, AI techniques to process data from sensors can detect patterns in physiological variables as well as positional and kinematic data to inform how athletes can improve their performance. Although AI has promising applications in sports medicine, there are several challenges that can hinder their adoption. We have also identified avenues for future work that can provide solutions to overcome these challenges.

Application of Improved VMD-LSTM Model in Sports Artificial Intelligence

In recent years, with the rapid development of a new generation of artificial intelligence technology, how to deeply apply artificial intelligence technology to physical education and break through the limitations of time-space scenarios and knowledge transfer methods in traditional models has become a key issue in intelligent physical education in the era of artificial intelligence. In order to realize the online monitoring of wearable devices with artificial intelligence in sports and overcome the problem of low recognition accuracy of electrocardiogram, blood oxygen, and respiratory signals in many cases, this paper proposes a combination of variational modal decomposition based on the maximum envelope kurtosis method. Long-short-term neural network (VMD-LSTM) monitoring method for wearable sports equipment. Through experimental analysis and verification, the current signal of the VMD model shows a trend of fluctuating from large to stable and then to large with motion, while the training accuracy of LSTM after the 150th iteration is 94.09%, which shows that the coupling model VMD LSTM can better predict the direction of sports artificial intelligence. In addition, although the training time of the BP neural network is shorter than that of the LSTM model, there is a large gap between the recognition effect and the LSTM, and there are also large differences between different neural network structures. This shows that the VMD-LSTM model has broad application prospects in such models.

Research on System Construction and Strategy of Intelligent Sports in the Implementation of National Fitness

This paper studies the construction and development strategy of intelligent sports system in the context of Chinese National Fitness Program with methods of literature review and model construction. The research shows that there are four dilemmas in the implementation of intelligent sports in national fitness: data security, market monopoly, legal supervision, and product iteration. However, there are also three promoting factors in this regard, including policy guarantee, market demand, and industrial upgrading. Following the principles of scientificity, effectiveness, public welfare, and collaboration, this paper designs a system for intelligent sports in national fitness. The construction of the national fitness intelligent sports system mainly consists of four modules, including basic framework construction, function design, content design, and operation analysis. With the systematic analysis of the status quo of intelligent sports application in national fitness, this paper puts forward intelligent sports development strategies in the implementation of national fitness from four aspects: optimizing the top-level design of government, speeding up industrial transformation and upgrading, constructing market supervision mechanism, and establishing a talent training system.

Artificial Intelligence and Machine Learning in Sport Research: An Introduction for Non-data Scientists

In the last two decades, artificial intelligence (AI) has transformed the way in which we consume and analyse sports. The role of AI in improving decision-making and forecasting in sports, amongst many other advantages, is rapidly expanding and gaining more attention in both the academic sector and the industry. Nonetheless, for many sports audiences, professionals and policy makers, who are not particularly au courant or experts in AI, the connexion between artificial intelligence and sports remains fuzzy. Likewise, for many, the motivations for adopting a machine learning (ML) paradigm in sports analytics are still either faint or unclear. In this perspective paper, we present a high-level, non-technical, overview of the machine learning paradigm that motivates its potential for enhancing sports (performance and business) analytics. We provide a summary of some relevant research literature on the areas in which artificial intelligence and machine learning have been applied to the sports industry and in sport research. Finally, we present some hypothetical scenarios of how AI and ML could shape the future of sports.

A Parametric Identification Method of Human Gait Differences and its Application in Rehabilitation

In order to understand the regularity of human motion, characteristic description is widely used in gait analysis. For completely expressing gait information and providing more concise indicators, parametric description is also particularly significant as a means of analysis. Therefore, in this paper, the mathematical models of gait curves based on the generalized extension-Bézier curve were investigated, of which the shape parameters were used as individual gait characteristics to distinguish whether the gait is normal or not and to assist in judging rehabilitation. To evaluate the models, angle data from three joints (hip, knee, and ankle) were recorded with motion capture system when participants (10 healthy males and 6 male patients with ankle fracture) were walking at comfortable velocity along a walkway. Then, the shape parameters of each subject were obtained by applying the mathematical models, and the parameter range of the normal group was further summarized. Through comparison, it could be found that most shape parameters of patients exceed the normal ranges in varying degrees, and are concentrated on specific parameters. The results can not only help to judge the recovery stages of patients but also figure out the corresponding abnormal postures, so as to provide guidance for rehabilitation training.

Current Approaches to the Use of Artificial Intelligence for Injury Risk Assessment and Performance Prediction in Team Sports: a Systematic Review

Background

The application of artificial intelligence (AI) opens an interesting perspective for predicting injury risk and performance in team sports. A better understanding of the techniques of AI employed and of the sports that are using AI is clearly warranted. The purpose of this study is to identify which AI approaches have been applied to investigate sport performance and injury risk and to find out which AI techniques each sport has been using.

Methods

Systematic searches through the PubMed, Scopus, and Web of Science online databases were conducted for articles reporting AI techniques or methods applied to team sports athletes.

Results

Fifty-eight studies were included in the review with 11 AI techniques or methods being applied in 12 team sports. Pooled sample consisted of 6456 participants (97% male, 25 ± 8 years old; 3% female, 21 ± 10 years old) with 76% of them being professional athletes. The AI techniques or methods most frequently used were artificial neural networks, decision tree classifier, support vector machine, and Markov process with good performance metrics for all of them. Soccer, basketball, handball, and volleyball were the team sports with more applications of AI.

Conclusions

The results of this review suggest a prevalent application of AI methods in team sports based on the number of published studies. The current state of development in the area proposes a promising future with regard to AI use in team sports. Further evaluation research based on prospective methods is warranted to establish the predictive performance of specific AI techniques and methods.

Electronic supplementary material

The online version of this article (10.1186/s40798-019-0202-3) contains supplementary material, which is available to authorized users.

Estimation of Ankle Joint Power during Walking Using Two Inertial Sensors

(1) Background: Ankle joint power, as an indicator of the ability to control lower limbs, is of great relevance for clinical diagnosis of gait impairment and control of lower limb prosthesis. However, the majority of available techniques for estimating joint power are based on inverse dynamics methods, which require performing a biomechanical analysis of the foot and using a highly instrumented environment to tune the parameters of the resulting biomechanical model. Such techniques are not generally applicable to real-world scenarios in which gait monitoring outside of the clinical setting is desired. This paper proposes a viable alternative to such techniques by using machine learning algorithms to estimate ankle joint power from data collected by two miniature inertial measurement units (IMUs) on the foot and shank, (2) Methods: Nine participants walked on a force-plate-instrumented treadmill wearing two IMUs. The data from the IMUs were processed to train and test a random forest model to estimate ankle joint power. The performance of the model was then evaluated by comparing the estimated power values to the reference values provided by the motion tracking system and the force-plate-instrumented treadmill. (3) Results: The proposed method achieved a high accuracy with the correlation coefficient, root mean square error, and normalized root mean square error of 0.98, 0.06 w/kg, and 1.05% in the intra-subject test, and 0.92, 0.13 w/kg, and 2.37% in inter-subject test, respectively. The difference between the predicted and true peak power values was 0.01 w/kg and 0.14 w/kg with a delay of 0.4% and 0.4% of gait cycle duration for the intra- and inter-subject testing, respectively. (4) Conclusions: The results of this study demonstrate the feasibility of using only two IMUs to estimate ankle joint power. The proposed technique provides a basis for developing a portable and compact gait monitoring system that can potentially offer monitoring and reporting on ankle joint power in real-time during activities of daily living.

A Wearable Gait Phase Detection System Based on Force Myography Techniques

(1) Background: Quantitative evaluation of gait parameters can provide useful information for constructing individuals’ gait profile, diagnosing gait abnormalities, and better planning of rehabilitation schemes to restore normal gait pattern. Objective determination of gait phases in a gait cycle is a key requirement in gait analysis applications; (2) Methods: In this study, the feasibility of using a force myography-based technique for a wearable gait phase detection system is explored. In this regard, a force myography band is developed and tested with nine participants walking on a treadmill. The collected force myography data are first examined sample-by-sample and classified into four phases using Linear Discriminant Analysis. The gait phase events are then detected from these classified samples using a set of supervisory rules; (3) Results: The results show that the force myography band can correctly detect more than 99.9% of gait phases with zero insertions and only four deletions over 12,965 gait phase segments. The average temporal error of gait phase detection is 55.2 ms, which translates into 2.1% error with respect to the corresponding labelled stride duration; (4) Conclusions: This proof-of-concept study demonstrates the feasibility of force myography techniques as viable solutions in developing wearable gait phase detection systems.

Effects of strengthening and stretching exercise programmes on kinematics and kinetics of running in older adults: a randomised controlled trial

The aim of this study was to investigate the effects of strengthening and stretching exercises on running kinematics and kinetics in older runners. One hundred and five runners (55-75 years) were randomly assigned to either a strengthening (n = 36), flexibility (n = 34) or control (n = 35) group. Running kinematics and kinetics were obtained using an eight-camera system and an instrumented treadmill before and after the eight-week exercise protocol. Measures of strength and flexibility were also obtained using a dynamometer and inclinometer/goniometer. A time effect was observed for the excursion angles of the ankle sagittal (P = 0.004, d = 0.17) and thorax/pelvis transverse (P < 0.001, d = 0.20) plane. Similarly, a time effect was observed for knee transverse plane impulse (P = 0.013, d = 0.26) and ground reaction force propulsion (P = 0.042, d = -0.15). A time effect for hip adduction (P = 0.006, d = 0.69), ankle dorsiflexion (P = 0.002, d = 0.47) and hip internal rotation (P = 0.048, d = 0.30) flexibility, and hip extensor (P = 0.001, d = -0.48) and ankle plantar flexor (P = 0.01, d = 0.39) strength were also observed. However, these changes were irrespective of exercise group. The results of the present study indicate that an eight-week stretching or strengthening protocol, compared to controls, was not effective in altering age-related running biomechanics despite changes in ankle and trunk kinematics, knee kinetics and ground reaction forces along with alterations in muscle strength and flexibility were observed over time.

Twenty-five years of sport performance research in the Journal of Sports Sciences

In this historical review covering the past 25 years, we reflect on the content of manuscripts relevant to the Sport Performance section of the Journal of Sports Sciences. Due to the wide diversity of sport performance research, the remit of the Sport Performance section has been broad and includes mathematical and statistical evaluation of competitive sports performances, match- and notation-analysis, talent identification, training and selection or team organization. In addition, due to the academic interests of its section editors, they adopted a quality-assurance role for the Sport Performance section, invariably communicated through key editorials that subsequently shaped the editorial policy of the Journal. Key high-impact manuscripts are discussed, providing readers with some insight into what might lead an article to become a citation "classic". Finally, landmark articles in the areas of "science and football" and "notation analysis" are highlighted, providing further insight into how such articles have contributed to the development of sport performance research in general and the Journal of Sports Sciences in particular.

Quantification of human motion: gait analysis-benefits and limitations to its application to clinical problems

The technology supporting the analysis of human motion has advanced dramatically. Past decades of locomotion research have provided us with significant knowledge about the accuracy of tests performed, the understanding of the process of human locomotion, and how clinical testing can be used to evaluate medical disorders and affect their treatment. Gait analysis is now recognized as clinically useful and financially reimbursable for some medical conditions. Yet, the routine clinical use of gait analysis has seen very limited growth. The issue of its clinical value is related to many factors, including the applicability of existing technology to addressing clinical problems; the limited use of such tests to address a wide variety of medical disorders; the manner in which gait laboratories are organized, tests are performed, and reports generated; and the clinical understanding and expectations of laboratory results. Clinical use is most hampered by the length of time and costs required for performing a study and interpreting it. A "gait" report is lengthy, its data are not well understood, and it includes a clinical interpretation, all of which do not occur with other clinical tests. Current biotechnology research is seeking to address these problems by creating techniques to capture data rapidly, accurately, and efficiently, and to interpret such data by an assortment of modeling, statistical, wave interpretation, and artificial intelligence methodologies. The success of such efforts rests on both our technical abilities and communication between engineers and clinicians.

Design and test of neural networks and statistical classifiers in computer-aided movement analysis: a case study on gait analysis

OBJECTIVE: To describe the methods for designing and testing diagnostic systems in movement analysis and to verify the clinical usefulness of neural networks and statistical classifiers in a case study. DESIGN: Connectionist and statistical models trained and tested with measured data. BACKGROUND: A basic need in rehabilitation and related fields is to efficiently manage the vast information obtained from a movement analysis laboratory. Many studies have dealt with the interpretation of measured variables in order to correlate objective descriptors to the presence and/or severity of specific neuromusculoskeletal disorders or their consequences. This traditional analytical approach has been complemented in the last decade by new non-linear classification tools called neural networks. METHODS: A gait analysis study on 148 lower limb arthrosis patients and 88 age-matched control subjects. Pathological and healthy gait patterns obtained from force plates wer discriminated by means of multilayer perceptrons and statistical classifiers. RESULTS: Ten input features were enough to train a multilayer perceptron with six hidden neurons. The discrimination rate of the neural net was 80% after cross-validation, significantly higher (P<0.05) than the performance of a Bayes quadratic classifier (about 75%). A great variance due to a small cross-validation set could be demonstrated. CONCLUSIONS: Strict statistical requirements must be observed for designing a neural network. Although these models attain a better performance than conventional statistical approaches, the benefits they bring are sometimes not sufficient to justify their use. Furthermore, clinicians routinely involved in critical decisions may not consider such diagnostic systems reliable enough.

Current issues in the mechanics of athletic activities. A position paper

This position paper reviews current research topics in the study of athletic activities, which biomechanists would consider to be important or contentious issues, and focuses on critically evaluating what needs to be done in future research. It concludes that there remain many unresolved issues in the mechanics of athletic activities, many of which overlap with other disciplines. These issues relate to injury mechanisms, the control and co-ordination of movement, and ways of providing biomechanical feedback to enhance performance in athletic activities. Research to address these important issues will increasingly become more question than discipline orientated, must focus more on mechanisms than description, and will involve teams of researchers interacting on interdisciplinary problems.