Artificial Intelligence in Sport: A Game-Changer for Injury Management? written by Luke Canavan Dignam.
Biography:
Luke is a first-year PhD student at Atlantic Technological University, Galway, researching the use of Artificial Intelligence to reduce the occurrence of sport-related concussions under the supervision of Ed Daly, Lisa Ryan, and Michael McCann. Luke graduated with a BSc in Sports and Exercise Science from Universidad Católica San Antonio de Murcia in July 2024, his dissertation focused on sport-related concussion. Since graduating, his passion for coaching has continued to grow through his work with Triathlon Ireland.
Blog:
Sports injuries are one of the most investigated domains in sports science, with organizations globally, exploring new strategies to reduce injuries, making it a priority for all stakeholders in sport. From athletes and physiotherapists to coaches, managers, and investors, are interested in how we approach injury prevention. Current seasonal structures put immense workload pressure on athletes, with injuries occurring exponentially across sports domains. Furthermore, injuries such as concussions and hamstring strains have been extensively researched, yet their prevalence remains largely unchanged. This has led my doctoral research into how we can use Artificial Intelligence (AI) to predict and reduce injury occurrences using readily available metrics (such as GPS data or wearables) in combination with skeletal tracking video systems to formulate a model that aids referees and key stakeholders in identifying an injury when it happens.
Skeletal tracking technology represents a breakthrough in injury detection and prevention. Analysing an athlete’s movement patterns in real-time, AI-driven skeletal tracking can detect abnormal biomechanical deviations that may indicate an injury or increased risk of one occurring. Unlike traditional wearables, which primarily provide external load metrics, skeletal tracking allows for a more precise analysis of joint angles, muscle activation patterns, and compensatory movements. This level of insight enables practitioners to intervene proactively, modifying workloads or adjusting movement mechanics before an injury manifests. It also aids referees and medical staff to make faster, data-backed decisions during live competition, ensuring that injuries are not overlooked or exacerbated by continued play.
Beyond real-time detection, AI is also transforming injury prediction. Machine learning models can analyse vast datasets, identifying risk factors based on historical injury records, workload trends, fatigue markers, and movement efficiency. These models allow practitioners to develop highly personalized injury risk profiles for each athlete, offering targeted interventions that go beyond generalized injury prevention strategies. By integrating multiple data streams—including GPS tracking, force plate analysis, and neuromuscular response tests—AI can predict when an athlete is approaching a dangerous threshold, enabling pre-emptive action to reduce injury likelihood. In elite sports, where the difference between peak performance and injury can be razor-thin, this predictive capability could redefine athlete management strategies.
The current buzz around AI is well-founded, It has been proven to reduce the workload for data analysts, medical practitioners, sports scientists, and others by automating many of the time-consuming tasks we previously spent extensive hours on. AI brings unparalleled power within risk analysis, performance metrics, insight delivery, and visualization to the sports practitioner. My journey into AI literacy began with the MSc in Digital Innovation in Sport at Atlantic Technological University (ATU), which provided a structured pathway from zero coding experience to developing my own supervised machine learning model. Through this course, I gained practical skills in Python and machine learning, allowing me to analyse real-world sports data and build predictive models for injury risk. This experience has shown me firsthand how AI literacy will become an essential part of every practitioner’s toolkit in the coming years, ensuring that sports professionals are not just consumers of AI-driven insights but active contributors to its development and application.
But AI is not without its challenges, there remains a reluctance among many organisations to adopt its use, and this skepticism should not be ignored or disregarded. The ongoing debate over data ownership—whether it belongs to the athlete or the organization—places AI at the centre of ethical scrutiny. High-profile AI models, such as OpenAI’s ChatGPT, have historically stored user data for future use, raising concerns about privacy and compliance with regulations like GDPR. This issue is particularly critical in youth sports, where data protection laws must be strictly upheld.
For now, we must recognize that AI has been developed by us, to be used by us, and will never replace the practitioner—especially in the sports industry. The human experience remains an irreplaceable factor in sports. AI will never fully comprehend the emotions of competition or the empathy of coaches, which gives us an undeniable edge. The gut feeling that practitioners develop over years of experience simply cannot be predicted by algorithms, and it is crucial that we respect this balance when integrating AI into sports.
The symbiotic relationship between AI and the practitioner will be the defining factor in separating organizations that lead in insight and innovation from those that fall behind. Practitioners who understand AI’s power and combine it with the human edge will deliver industry-leading advantages. Those who proactively integrate AI while honing their expertise in interpretation and decision-making will lead the future of sports innovation. AI should be seen not as a replacement, but as a powerful partner—one that enhances human expertise while preserving the essence of sport.