Closing Pandora's Box: adapting a systems ergonomics methodology for better understanding the ecological complexity underpinning the development and prevention of running-related injury

Artificial intelligence and Machine Learning approaches in sports: Concepts, applications, challenges, and future perspectives

BACKGROUND

The development and application of Artificial Intelligence (AI) and Machine Learning (ML) in healthcare have gained attention as a promising and powerful resource to change the landscape of healthcare. The potential of these technologies for injury prediction, performance analysis, personalized training, and treatment comes with challenges related to the complexity of sports dynamics and the multidimensional aspects of athletic performance.

OBJECTIVES

We aimed to present the current state of AI and ML applications in sports science, specifically in the areas of injury prediction, performance enhancement, and rehabilitation. We also examine the challenges of incorporating AI and ML into sports and suggest directions for future research.

METHOD

We conducted a comprehensive literature review, focusing on publications related to AI and ML applications in sports. This review encompassed studies on injury prediction, performance analysis, and personalized training, emphasizing the AI and ML models applied in sports.

RESULTS

The findings highlight significant advancements in injury prediction accuracy, performance analysis precision, and the customization of training programs through AI and ML. However, future studies need to address challenges such as ethical considerations, data quality, interpretability of ML models, and the integration of complex data.

CONCLUSION

AI and ML may be useful for the prevention, detection, diagnosis, and treatment of health conditions. In this Masterclass paper, we introduce AI and ML concepts, outline recent breakthroughs in AI technologies and their applications, identify the challenges for further progress of AI systems, and discuss ethical issues, clinical and research opportunities, and future perspectives.

Factors Influencing Runner's Choices of Footwear

Until the mid-2000s, qualitative research has been virtually absent from running injury research. A handful of studies have been recently published regarding the attitudes and perceptions of runners and coaches toward injury development. Footwear is frequently perceived as a risk factor for running related injuries, but empeirical evidence fails to support such beliefs. The reasons why runners choose specific footwear warrants formal investigation to further understand the links between footwear and running related injuries. This study aimed to investigate the factors influencing runners choices of footwear. Interviews were conducted with 12 runners. Recordings from the interviews were transcribed verbatim and themes were developed using thematic analysis. Findings revealed 15 unique factors that influence runner's choices of footwear for running. These factors were grouped into three main themes: personal footwear characteristic preferences, other people and economic considerations. Runners largely gather information about their footwear choices from past experiences and people they trust and admire. They also emphasized the complexity of footwear choices due to availability and the constant changes preset within the footwear industry. This research adds to the growing body of knowledge to better understand the wider running injury system. Further studies are needed to establish how runners perceptions of their footwear impact injury rates and to develop effective injury prevention strategies.

Systems thinking-based risk assessment methods applied to sports performance: A comparison of STPA, EAST-BL, and Net-HARMS in the context of elite women's road cycling

There is increasing interest in applying systems Human Factors and Ergonomics (HFE) methods in sport. Risk assessment (RA) methods can be used identify risks which may impact the performance of individual athletes, teams, and overall sports systems; however, they have not yet been tested in sport. This study sets out to apply and compare three systems thinking-based RA methods in the context of elite sports performance and report on the frequency and types of the risks identified. The Systems-Theoretic Process Analysis (STPA) method, the Event Analysis of Systemic Teamwork Broken Links (EAST-BL) method, and the Networked Hazard Analysis and Risk Management System (Net-HARMS) method were applied to elite women's road cycling to identify all the credible risks that could degrade optimal team performance. The findings demonstrate that all three methods appear to provide useful results in a context other than safety, and that multiple risks threatening the performance of the cycling team were identified. Whilst the frequency and types of risks differed across the methods applied, there are additional theoretical, methodological, and practical implications to be considered prior to the selection and use of systems thinking-based RA approaches. Recommendations and directions for future HFE and sports science research are discussed.

Out of control? Using STAMP to model the control and feedback mechanisms surrounding identity crime in darknet marketplaces

Darknet marketplaces have emerged as a facilitator of identity crime and trading. This study aimed to (1) understand the entities and control and feedback mechanisms that influence identity crime prevention and occurrence on the darknet in the Australian system and to (2) comprehensively identify the implications of control failures across all system levels. The Systems-Theoretic Accident Model and Processes (STAMP) was used to develop an identity crime control structure in consultation with subject matter experts and then the Systems-Theoretic Process Analysis (STPA) was applied. The STPA identified 310 risk states, resulting from control failures and which were associated with the range of agencies, organisations, and individuals present across system levels. As darknet marketplaces rapidly evolve, alignment between these entities is necessary to enable agile system responses. STAMP and STPA have promise in understanding the potential for intervention across all system levels in preventing societal issues such as identity crime.

STPA methodology in a socio-technical system of monitoring and tracking diabetes mellitus

In this paper, an adaptation of the system-theoretic process analysis (STPA) approach to a sociotechnical system of monitoring and tracking diabetes mellitus (DM) is proposed. Several human controllers such as the patient, physicians, other healthcare providers, and people in the social environment were considered in the hierarchical safety control structure. During the identification of the causal factors (causal scenarios) for the unsafe control actions of the human controllers, the three-phase approach proposed by France (2017) as well as the interaction problems between the subsystems in the hierarchical safety control structure were considered, and the implementation of the basic safety design procedure was included in the methodology for the improvement of system. A dynamic STPA methodology is also proposed as an extension to the basic methodology.

Training Load and Its Role in Injury Prevention, Part I: Back to the Future

The purpose of this 2-part commentary series is† to explain why we believe our ability to control injury risk by manipulating training load (TL) in its current state is an illusion and why the foundations of this illusion are weak and unreliable. In part 1, we introduce the training process framework and contextualize the role of TL monitoring in the injury-prevention paradigm. In part 2, we describe the conceptual and methodologic pitfalls of previous authors who associated TL and injury in ways that limited their suitability for the derivation of practical recommendations. The first important step in the training process is developing the training program: the practitioner develops a strategy based on available evidence, professional knowledge, and experience. For decades, exercise strategies have been based on the fundamental training principles of overload and progression. Training-load monitoring allows the practitioner to determine whether athletes have completed training as planned and how they have coped with the physical stress. Training load and its associated metrics cannot provide a quantitative indication of whether particular load progressions will increase or decrease the injury risk, given the nature of previous studies (descriptive and at best predictive) and their methodologic weaknesses. The overreliance on TL has moved the attention away from the multifactorial nature of injury and the roles of other important contextual factors. We argue that no evidence supports the quantitative use of TL data to manipulate future training with the purpose of preventing injury. Therefore, determining "how much is too much" and how to properly manipulate and progress TL are currently subjective decisions based on generic training principles and our experience of adjusting training according to an individual athlete's response. Our message to practitioners is to stop seeking overly simplistic solutions to complex problems and instead embrace the risks and uncertainty inherent in the training process and injury prevention.

Sports-related concussion management as a control problem: using STAMP to examine concussion management in community rugby

The effective management of sports-related concussion is an on-going problem in amateur sport due to a number of systemic issues. These factors have often been studied in isolation with minimal consideration for the interactions between them, or the overall system in which they occur. The aim of this research was to model the actors, controls, and feedback mechanisms influencing the management of concussion in community rugby union using the Systems-Theoretic Accident Model and Processes (STAMP) method. Findings show that there are currently many inadequate controls, that are directly and indirectly impacting effective concussion management. The practical implications of these findings are discussed (e.g. improved guideline education, mandatory medical presence, rule amendments). Additionally, the model provides a sound framework similar team sports can use to inform research into injury management and prevention (e.g. rugby league, hockey, AFL). Practitioner Summary: Research shows that concussion management in regional sport is inconsistent due to contextual limitations. Systems-Theoretic Accident Model and Processes was applied to identify the systemic factors currently influencing concussion management practices in community sport. Findings show inadequate controls precipitate concussion management gaps. Practical implications of the findings are discussed.

Applying systems ergonomics methods in sport: A systematic review

INTRODUCTION

As sports systems become increasingly more complex, competitive, and technology-centric, there is a greater need for systems ergonomics methods to consider the performance, health, and safety of athletes in context with the wider settings in which they operate. Therefore, the purpose of this systematic review was to identify and critically evaluate studies which have applied a systems ergonomics research approach in the context of sports performance and injury management.

MATERIAL AND METHODS

Five databases (PubMed, Scopus, ScienceDirect, Web of Science, and SPORTDiscus) were searched for the dates 01 January 1990 to 01 August 2017, inclusive, for original peer-reviewed journal articles and conference papers. Reported analyses were underpinned by a recognised systems ergonomics method, and study aims were related to the optimisation of sports performance (e.g. communication, playing style, technique, tactics, or equipment), and/or the management of sports injury (i.e. identification, prevention, or treatment).

RESULTS

A total of seven articles were identified. Two articles were focussed on understanding and optimising sports performance, whereas five examined sports injury management. The methods used were the Event Analysis of Systemic Teamwork, Cognitive Work Analysis (the Work Domain Analysis Abstraction Hierarchy), Rasmussen's Risk Management Framework, and the Systems Theoretic Accident Model and Processes method. The individual sport application was distance running, whereas the team sports contexts examined were cycling, football, Australian Football League, and rugby union.

CONCLUSIONS

The included systems ergonomics applications were highly flexible, covering both amateur and elite sports contexts. The studies were rated as valuable, providing descriptions of injury controls and causation, the factors influencing injury management, the allocation of responsibilities for injury prevention, as well as the factors and their interactions underpinning sports performance. Implications and future directions for research are described.

Towards a complex systems approach in sports injury research: simulating running-related injury development with agent-based modelling

OBJECTIVES

There have been recent calls for the application of the complex systems approach in sports injury research. However, beyond theoretical description and static models of complexity, little progress has been made towards formalising this approach in way that is practical to sports injury scientists and clinicians. Therefore, our objective was to use a computational modelling method and develop a dynamic simulation in sports injury research.

METHODS

Agent-based modelling (ABM) was used to model the occurrence of sports injury in a synthetic athlete population. The ABM was developed based on sports injury causal frameworks and was applied in the context of distance running-related injury (RRI). Using the acute:chronic workload ratio (ACWR), we simulated the dynamic relationship between changes in weekly running distance and RRI through the manipulation of various 'athlete management tools'.

RESULTS

The findings confirmed that building weekly running distances over time, even within the reported ACWR 'sweet spot', will eventually result in RRI as athletes reach and surpass their individual physical workload limits. Introducing training-related error into the simulation and the modelling of a 'hard ceiling' dynamic resulted in a higher RRI incidence proportion across the population at higher absolute workloads.

CONCLUSIONS

The presented simulation offers a practical starting point to further apply more sophisticated computational models that can account for the complex nature of sports injury aetiology. Alongside traditional forms of scientific inquiry, the use of ABM and other simulation-based techniques could be considered as a complementary and alternative methodological approach in sports injury research.

A Systems Approach to Performance Analysis in Women's Netball: Using Work Domain Analysis to Model Elite Netball Performance

Netball is a newly professional women’s sport, as such there has been little research conducted investigating performance analysis (PA) in elite netball. The aim of this study was to develop a model of the elite netball performance system to identify the complex relationships among key performance indicators. Eleven elite subject matter experts (SMEs) participated in workshops to produce a systems model of the netball match performance. The model was developed using the work domain analysis (WDA) method. A model of the netball match performance system was produced showing the interrelated objects, processes, functions, values, and purposes involved in elite level netball matches. The model identified the components of elite level netball performance and the interactions and relationships between them. The output of this research has identified novel PA measures including passing and possession measures, measures of cognitive performance, and measures related to physical activity. Netball is a complex sport, involving multiple dynamic and interrelated components. Consequently, there is an opportunity to develop holistic PA measures that focus on interacting components, as opposed to components in isolation.

Getting the most out of intensive longitudinal data: a methodological review of workload-injury studies

OBJECTIVES

To systematically identify and qualitatively review the statistical approaches used in prospective cohort studies of team sports that reported intensive longitudinal data (ILD) (>20 observations per athlete) and examined the relationship between athletic workloads and injuries. Since longitudinal research can be improved by aligning the (1) theoretical model, (2) temporal design and (3) statistical approach, we reviewed the statistical approaches used in these studies to evaluate how closely they aligned these three components.

DESIGN

Methodological review.

METHODS

After finding 6 systematic reviews and 1 consensus statement in our systematic search, we extracted 34 original prospective cohort studies of team sports that reported ILD (>20 observations per athlete) and examined the relationship between athletic workloads and injuries. Using Professor Linda Collins' three-part framework of aligning the theoretical model, temporal design and statistical approach, we qualitatively assessed how well the statistical approaches aligned with the intensive longitudinal nature of the data, and with the underlying theoretical model. Finally, we discussed the implications of each statistical approach and provide recommendations for future research.

RESULTS

Statistical methods such as correlations, t-tests and simple linear/logistic regression were commonly used. However, these methods did not adequately address the (1) themes of theoretical models underlying workloads and injury, nor the (2) temporal design challenges (ILD). Although time-to-event analyses (eg, Cox proportional hazards and frailty models) and multilevel modelling are better-suited for ILD, these were used in fewer than a 10% of the studies (n=3).

CONCLUSIONS

Rapidly accelerating availability of ILD is the norm in many fields of healthcare delivery and thus health research. These data present an opportunity to better address research questions, especially when appropriate statistical analyses are chosen.

SeRUN® study: Development of running profiles using a mixed methods analysis

OBJECTIVE

To determine profiles of urban runners based on socio-demographic, health, motivational, training characteristics and running-related beliefs and behaviours.

METHODS

Mixed, exploratory, sequential study with two stages: 1) quantitative, using an online survey; and 2) qualitative, using semi-structured interviews with runners from the previous stage. Participants were recruited via: running routes commonly attended by runners, eight races, previous databases and social media networks. The survey collected information on six dimensions: (1) socio-demographic; (2) health; (3) motivations; (4) training characteristics; (5) running-related behaviour; and (6) beliefs and perceptions about health. Profiles were identified using a two-step hierarchical clustering analysis. Subsequently, 15 interviews were conducted with participating runners across each of the identified profiles. Qualitative analysis complemented the profiles characterization, explaining motivations to start and continue running, beliefs about risk factors and injury prevention, and the physical therapist's role in rehabilitation. Statistical analysis from stage one was conducted using SPSS 22 with a confidence level of 5%. Qualitative data were analysed using thematic and content analyses.

RESULTS

A total of 821 surveys were analysed (46% female), mean aged 36.6±10.0 years. Cluster analysis delineated four profiles (n = 752) according to years of running experience, weekly running volume and hours of weekly training. Profiles were named "Beginner" (n = 163); "Basic" (n = 164); "Middle" (n = 160) and "Advanced" (n = 265). Profiles were statistically different according to sex, age, years of running experience, training characteristics, previous injuries and use of technological devices (p<0.05). There were identified motivations to start and continue running. Beliefs about risk factors vary among stretching, footwear, training surface and overload. Runners identified the physical therapist as a specialist, involved in the rehabilitation process and showing empathy towards the patient. The identification of these profiles allows the generation of future prospective studies and clinical trials to evaluate risk and prognostic factors targeting specific populations of runners, with the ultimate aim of reducing running-related injury.

From control to causation: Validating a 'complex systems model' of running-related injury development and prevention

INTRODUCTION

There is a need for an ecological and complex systems approach for better understanding the development and prevention of running-related injury (RRI). In a previous article, we proposed a prototype model of the Australian recreational distance running system which was based on the Systems Theoretic Accident Mapping and Processes (STAMP) method. That model included the influence of political, organisational, managerial, and sociocultural determinants alongside individual-level factors in relation to RRI development. The purpose of this study was to validate that prototype model by drawing on the expertise of both systems thinking and distance running experts.

MATERIALS AND METHODS

This study used a modified Delphi technique involving a series of online surveys (December 2016- March 2017). The initial survey was divided into four sections containing a total of seven questions pertaining to different features associated with the prototype model. Consensus in opinion about the validity of the prototype model was reached when the number of experts who agreed or disagreed with survey statement was ≥75% of the total number of respondents.

RESULTS

A total of two Delphi rounds was needed to validate the prototype model. Out of a total of 51 experts who were initially contacted, 50.9% (n = 26) completed the first round of the Delphi, and 92.3% (n = 24) of those in the first round participated in the second. Most of the 24 full participants considered themselves to be a running expert (66.7%), and approximately a third indicated their expertise as a systems thinker (33.3%). After the second round, 91.7% of the experts agreed that the prototype model was a valid description of the Australian distance running system.

CONCLUSION

This is the first study to formally examine the development and prevention of RRI from an ecological and complex systems perspective. The validated model of the Australian distance running system facilitates theoretical advancement in terms of identifying practical system-wide opportunities for the implementation of sustainable RRI prevention interventions. This 'big picture' perspective represents the first step required when thinking about the range of contributory causal factors that affect other system elements, as well as runners' behaviours in relation to RRI risk.