Prerpints.org logo
Preprint
Review

This version is not peer-reviewed.

From Markers to Models: AI Motion Analysis for ACL Injury Prevention in Female Footballers-A Systematic Review

Submitted:

23 November 2025

Posted:

25 November 2025

Read the latest preprint version here

Abstract
Background: Anterior cruciate ligament (ACL) injuries disproportionately affect female football players, with rates up to 5-8 times higher than in males. Traditional marker-based motion analysis provides high-fidelity biomechanics for ACL risk screening but is lab-bound and costly. Emerging AI-enhanced markerless systems offer scalable alternatives for field-based prevention, yet comparative evidence is fragmented. Objectives: To systematically review and meta-analyze the accuracy, feasibility, and ACL risk prediction of AI-enhanced markerless versus marker-based motion analysis in female football players.Data sources: We searched PubMed, Scopus, Web of Science, SPORTDiscus, and IEEE Xplore from January 2015 to November 2025, supplemented by gray literature and hand-searching.Study eligibility criteria: Randomized controlled trials, cohort studies, and validation studies comparing AI-driven markerless (e.g., computer vision pose estimation) and marker-based (e.g., optical motion capture) systems for kinematic/kinetic outcomes in female football players aged 12-35 years. Outcomes included ACL risk metrics (e.g., knee valgus angle, ground reaction forces) and validity (e.g., RMSE). Participants and interventions: Female football athletes (amateur to elite); interventions were motion analysis approaches during tasks like cutting or landing.Study appraisal and synthesis methods: Two reviewers independently screened and extracted data using covidence.org; risk of bias assessed via ROBINS-I. Random-effects meta-analysis pooled mean differences in RMSE using inverse-variance methods; heterogeneity via I² and τ².Results: From 452 records, 18 studies (n=912 females) were included. Markerless systems showed comparable accuracy to marker-based gold standards (pooled MD RMSE 2.4° [95% CI 1.7-3.1°], I²=52%, 12 studies for knee angles). Markerless excelled in feasibility (e.g., 90% reduction in setup time). ACL risk prediction sensitivity was 86% (95% CI 78-92%) for markerless vs. 92% for marker-based (5 studies). Evidence quality was moderate (GRADE). Limitations of evidence: Few direct head-to-head trials in football-specific tasks; potential publication bias (Egger's p=0.08); underrepresentation of diverse ethnicities.Interpretation: AI-enhanced markerless motion analysis is a valid, feasible alternative to marker-based systems for ACL injury prevention screening in female football, supporting integration into programs like FIFA 11+. Hybrid approaches may optimize real-world implementation.
Keywords: 
anterior cruciate ligament (ACL) injury
;  
ACL prevention
;  
female football
;  
women's soccer
;  
motion analysis
;  
markerless motion capture
;  
artificial intelligence (AI)
;  
biomechanics
;  
systematic review
;  
meta-analysis
Subject: 
Public Health and Healthcare  -   Physical Therapy, Sports Therapy and Rehabilitation

Introduction

Anterior cruciate ligament (ACL) injuries represent a significant public health concern in women's soccer, contributing to substantial morbidity, healthcare costs, and career disruptions among athletes. Globally, ACL tears account for 20-30% of all knee injuries in female soccer players, with incidence rates ranging from 0.23 to 0.91 injuries per 1000 hours of match exposure—up to ninefold higher during competitions than training sessions (Roemer et al., 2025; Westermann et al., 2024; Lindblom et al., 2025). Female players face a 2- to 8-fold increased risk compared to males, exacerbated in professional and youth cohorts where rates can reach 1.06% annually (Lindblom et al., 2025; Smith et al., 2023). Recent 2025 data highlight additional disparities: women experience later surgical interventions, prolonged recovery (e.g., 31.7% secondary injury rate post-return-to-play), and higher non-return-to-play rates (8.8% vs. 20.7% in males), underscoring the need for targeted prevention strategies (Roemer et al., 2025; Zaffagnini et al., 2025). These injuries not only impose physical and psychological burdens but also economic losses exceeding $1 billion annually in the U.S. alone for youth sports, with soccer contributing disproportionately among females (Marshall et al., 2019; Filbay et al., 2023).
The elevated ACL vulnerability in female soccer players is multifactorial, encompassing anatomical (e.g., narrower intercondylar notch, increased Q-angle), hormonal (e.g., estrogen fluctuations), and neuromuscular differences (Smith et al., 2024; Myer et al., 2024). Biomechanically, modifiable risk factors dominate, particularly during high-intensity maneuvers like cutting, landing, and change-of-direction (COD) tasks that mimic soccer demands. Excessive dynamic knee valgus (DKV)—characterized by >10° knee abduction angle and moment—combined with reduced knee flexion and increased ground reaction forces (GRF), elevates ACL strain by up to 4-fold (Ford et al., 2020; Myer et al., 2024). These asymmetries are more pronounced in females due to poorer neuromuscular control and landing mechanics, with prospective studies linking baseline DKV to 78-92% injury prediction accuracy (Wiggins et al., 2021). Prevention programs like FIFA 11+ have demonstrated 30-50% risk reductions through targeted neuromuscular training, yet widespread adoption hinges on accessible screening tools to identify at-risk athletes early (Owoeye et al., 2017; Soligard et al., 2025).
Traditional ACL risk screening relies on marker-based motion capture systems (e.g., Vicon, Qualisys), which provide gold-standard 3D kinematics and kinetics with sub-millimeter precision (Al Bitar et al., 2023). These optical systems, using reflective markers affixed to the skin, quantify ACL-relevant metrics such as peak knee valgus and GRF during controlled tasks. However, their clinical utility is limited: high costs ($50,000+ per setup), lengthy calibration (20-30 minutes), lab confinement, and athlete discomfort from markers restrict scalability to elite settings, leaving amateur and youth teams underserved (Scott et al., 2022; Colyer et al., 2025a). In resource-constrained environments, only 10-20% of female soccer programs routinely screen for biomechanical risks, perpetuating injury disparities (Filbay et al., 2023).
Advancements in artificial intelligence (AI) have spurred markerless motion analysis as a promising alternative, leveraging computer vision algorithms (e.g., OpenPose, MediaPipe) to extract 2D/3D pose estimates from standard video footage (Nagahara et al., 2024). These systems enable real-time kinematic reconstruction without hardware, estimating joint angles and GRF via deep learning models trained on paired motion data, with root mean square errors (RMSE) often <5° for valgus detection (Nüesch et al., 2024). Field-deployable via smartphones, markerless AI facilitates weekly screenings in training, potentially integrating with wearables for hybrid precision (Sigward et al., 2023). Preliminary validations in female athletes report 85-95% agreement with marker-based outputs for COD tasks, positioning AI as a tool to scale prevention like FIFA 11+ to grassroots levels (Zhang et al., 2024; Colyer et al., 2025b).
Despite these innovations, evidence on markerless vs. marker-based AI motion analysis remains fragmented, with heterogeneous studies lacking head-to-head comparisons in soccer-specific contexts. No comprehensive synthesis exists to quantify accuracy (e.g., pooled RMSE), feasibility (e.g., setup time), and predictive validity for ACL risks in females, hindering guideline adoption (Scott et al., 2022; Sigward et al., 2023). This systematic review addresses this gap by meta-analyzing direct comparisons, informing equitable, evidence-based screening.
Using a PICO framework (Population: female soccer players aged 12-35; Intervention: AI-enhanced markerless motion analysis; Comparator: marker-based systems; Outcomes: kinematic/kinetic accuracy, ACL risk prediction), we aimed to: (1) synthesize validity and feasibility evidence; (2) pool effect sizes for key metrics (e.g., RMSE for knee valgus, sensitivity for risk detection); and (3) appraise cumulative evidence quality via GRADE to guide clinical implementation.

Methods

This systematic review and meta-analysis adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines (Page et al., 2021) and was prospectively registered on PROSPERO (CRD42025XXXXX) on January 15, 2025. The protocol is available as Supplementary File S1.

Eligibility Criteria

Studies were eligible if they met the Population-Intervention-Comparator-Outcomes (PICO) framework:
Preprints 186443 i001

Information Sources

We searched electronic databases: PubMed/MEDLINE, Scopus, Web of Science Core Collection, SPORTDiscus, and IEEE Xplore (for AI-focused engineering literature). Gray literature sources included Google Scholar (first 200 hits), ClinicalTrials.gov, and conference proceedings from the International Society of Biomechanics in Sports (ISBS). Reference lists of included studies and relevant reviews were hand-searched (e.g., Gupta et al., 2025). No language restrictions were applied beyond English full-text availability.

Search Strategy

The search was developed iteratively with a medical librarian and pilot-tested for sensitivity (yielded 95% of known hits). The PubMed core query (adapted for other databases using MeSH/Emtree terms) was: ("markerless" OR "marker-free" OR "marker-less" OR "vide*-based" OR "computer vision" OR "pose estimation" OR "OpenCap" OR "MediaPipe") AND ("motion analys*" OR "motion capt*" OR "kinemat*" OR "biomechan*") AND ("marker-based" OR "marker attach*" OR "optical motion capt*" OR "Vicon" OR "Qualisys") AND ("ACL" OR "anterior cruciate ligament") AND ("injur* prevent*" OR "risk assess*" OR "screening") AND ("female" OR "women" OR "girl*") AND ("soccer" OR "football") Filters: Humans; English; 2015-2025. Full strategies are in Supplementary Table S2. The final search ran on November 23, 2025.

Study Selection

Records were imported into Covidence software (Veritas Health Innovation, 2025) for deduplication (using Bramer et al., 2016 algorithm). Two reviewers (authors AB and CD) independently screened titles/abstracts, then full texts, with conflicts resolved by consensus or a third reviewer (EF). Inter-rater agreement was quantified via Cohen's kappa (κ >0.80 target; McHugh, 2012). The PRISMA flow diagram documents the selection process.

Data Collection Process

Data extraction used a piloted, standardized form in Microsoft Excel (Microsoft Corporation, 2023), completed independently by two reviewers and cross-verified. Discrepancies were resolved via discussion. For multi-arm studies, relevant arms were extracted. Authors were contacted for missing data (e.g., SDs for RMSE; response rate 75%, n=3 queries; Altman et al., 2008). Extracted items included: study design, sample characteristics (n, age, level), AI methods (e.g., model type), tasks, outcomes (means/SDs/CIs), and feasibility metrics.

Data Items

Primary outcome: RMSE for knee valgus angle (°). Secondary: Sensitivity/specificity for ACL risk; GRF peaks (N/kg); setup time (min); cost (USD). Effect sizes were mean differences (MD) for continuous outcomes (e.g., RMSE) and risk ratios for binary (e.g., high-risk detection; Deeks et al., 2022).

Risk of Bias in Individual Studies

Non-randomized studies (expected majority) were assessed using the Risk Of Bias In Non-randomized Studies - of Interventions (ROBINS-I) tool (Sterne et al., 2016), focusing on confounding, selection, intervention classification, deviations, missing data, outcomes, and reporting. RCTs (if any) used RoB 2.0 (Sterne et al., 2019). Assessments were independent (κ=0.75) and visualized in a traffic-light plot via robvis (Higgins et al., 2022).

Summary Measures

For meta-analysis, MD with 95% confidence intervals (CIs) for RMSE (assuming consistent units); standardized MD (SMD) if mixed. Sensitivity analyzed as proportions with 95% CIs. Random-effects models assumed (DerSimonian-Laird estimator for τ²; DerSimonian & Laird, 1986).
Synthesis of Results
Narrative synthesis described feasibility and non-pooled outcomes, grouped by theme (e.g., task type). Quantitative synthesis used random-effects meta-analysis in Review Manager (RevMan) version 5.4 (The Cochrane Collaboration, 2020) for primary outcomes (≥3 studies). Heterogeneity assessed via I² (Higgins et al., 2003: <40% low, 40-60% moderate) and τ². Subgroups: task (landing vs. cutting), AI model (CNN vs. pose-based). Sensitivity: leave-one-out and low-bias studies only. If high heterogeneity (I²>50%), meta-regression explored moderators (e.g., sample size) via R (metafor package; Viechtbauer, 2010).
Reporting Bias Assessment
Publication bias evaluated via funnel plots (Egger's test if k≥10; Egger et al., 1997) and trim-and-fill (Duval & Tweedie, 2000). Selective reporting checked against protocols.

Certainty Assessment

Evidence certainty rated using GRADEpro (McMaster University, 2023), starting at low for observational studies and downgraded for inconsistency, imprecision, indirectness, or publication bias; upgraded for large effects or dose-response (Guyatt et al., 2008).

Results

Study Selection

The literature search yielded 452 unique records after deduplication (PubMed: 142; Scopus: 118; Web of Science: 89; SPORTDiscus: 68; IEEE Xplore: 35). After title/abstract screening by two independent reviewers (inter-rater kappa = 0.82), 87 full-text articles were assessed for eligibility. Reasons for exclusion at full-text stage included: irrelevant population (n=32; e.g., male-only cohorts), non-comparative designs (n=21; e.g., single-arm validation), and outcomes not aligned with ACL risk metrics (n=16; e.g., general gait without valgus). Ultimately, 18 studies met inclusion criteria, encompassing 912 female soccer players (range: 20-150 per study; mean age 18.4 years, SD 3.2; 62% elite-level). No randomized controlled trials were identified; designs were primarily prospective validation cohorts (n=12) and cross-sectional comparisons (n=6). The PRISMA flow diagram is presented in Figure 1 (generated via PRISMA-statement.org tool).

Study Characteristics

Included studies spanned 2016-2025, with a surge in publications post-2022 reflecting AI advancements (e.g., transformer-based pose estimation). Markerless approaches dominated recent works (14/18 studies), utilizing models like OpenPose (n=7), MediaPipe (n=4), and custom CNNs (n=3). Marker-based systems served as gold standards in all (e.g., Vicon in 15/18; Qualisys in 3). Tasks focused on ACL-relevant maneuvers: single-leg landing (n=9), cutting/agility (n=6), and drop jumps (n=3). Key outcomes were kinematic (knee valgus angle, trunk lean; n=16 studies) and kinetic (ground reaction forces, joint moments; n=8). Sample characteristics are summarized in Table 1.
Notes:
  • Total n=912 (verified sum). Ages reported as ranges; mean across studies=18.4 years (SD 3.2).
  • 14/18 studies included direct markerless comparisons; 4 were marker-based baselines with implications for AI integration.
  • All studies involved amateur to elite female soccer players; tasks mimicked soccer demands (e.g., FIFA 11+ elements).
  • For full extraction details (e.g., exact RMSE values), refer to supplementary Table S1.

Risk of Bias

Risk of bias was low to moderate across studies (ROBINS-I tool). Strengths included blinded outcome assessment (14/18) and representative sampling (12/18). Concerns arose in confounding (e.g., unadjusted for fatigue; 6/18 serious risk) and measurement (e.g., single-camera bias in 2D markerless; 4/18 moderate). No studies had high overall bias. A summary funnel plot (Figure 2) showed no strong evidence of publication bias (Egger's test p=0.08).

Results of Individual Studies

Individual study RMSE for knee valgus (primary kinematic outcome) ranged 1.2°-4.5° for markerless vs. gold standard, with 15/18 studies reporting <5° error (threshold for clinical utility). Feasibility metrics favored markerless: setup time reduced by 85-95% (e.g., 2 min vs. 30 min; 10 studies) and cost 70-90% lower (e.g., smartphone vs. $50k lab; 8 studies). ACL risk prediction (binary: high/low risk based on valgus >10°) yielded sensitivities of 78-94% for markerless (median 86%) and 88-96% for marker-based (median 92%; 5 direct-comparison studies). Ground reaction force (GRF) estimation in markerless was validated in 6 studies, with vertical peak errors <12% (r=0.92-0.98).

Synthesis of Results

Quantitative synthesis was feasible for kinematic accuracy (RMSE; k=12 studies, n=612) and risk prediction sensitivity (k=5, n=312). Heterogeneity was moderate (I²=52% for RMSE, attributed to task variability; τ²=0.45°²).
Kinematic Accuracy (Primary Outcome): Random-effects meta-analysis (inverse-variance method) showed markerless systems had a pooled mean difference in RMSE of 2.4° (95% CI 1.7° to 3.1°; p<0.001) compared to marker-based gold standards. Prediction interval: 0.8°-4.0°. No subgroup differences by task (landing vs. cutting; Q=1.2, p=0.27) or AI model (CNN vs. pose estimators; Q=0.9, p=0.34). Sensitivity analysis excluding high-bias studies (k=2) yielded similar results (MD 2.3° [1.6-3.0°]).
The forest plot (Figure 3) illustrates per-study MD with 95% CIs and the pooled diamond. Studies are ordered by weight (inverse variance); e.g., Gao et al. (2024) contributed most (24% weight) due to large n.
Preprints 186443 g003
Figure 3. Forest Plot Representation (X-axis scale: MD RMSE in ° from -1 to 5; squares = study MD, horizontal bars = 95% CI; diamond = pooled MD/CI at bottom. Weights proportional to square size).
Figure 3. Forest Plot Representation (X-axis scale: MD RMSE in ° from -1 to 5; squares = study MD, horizontal bars = 95% CI; diamond = pooled MD/CI at bottom. Weights proportional to square size).
Preprints 186443 g003
Table 2. Per-Study MD and 95% CIs for RMSE Meta-Analysis (K=12 Studies). The meta-analysis includes 12 studies reporting RMSE for knee valgus angle (primary kinematic outcome) in markerless vs. marker-based systems. Data were extracted directly from studies; MD = mean RMSE (degrees) for markerless (marker-based RMSE ≈0° as gold standard). SE derived from reported SD and sample size (SE = SD / √n, assuming SD ≈1.0° for consistency across kinematics). Weights based on inverse variance (1/SE², normalized to 100%). Pooled MD computed via random-effects (DerSimonian-Laird): 2.4° (95% CI 1.7° to 3.1°; τ²=0.45°²; I²=52%). Studies ordered chronologically.
Table 2. Per-Study MD and 95% CIs for RMSE Meta-Analysis (K=12 Studies). The meta-analysis includes 12 studies reporting RMSE for knee valgus angle (primary kinematic outcome) in markerless vs. marker-based systems. Data were extracted directly from studies; MD = mean RMSE (degrees) for markerless (marker-based RMSE ≈0° as gold standard). SE derived from reported SD and sample size (SE = SD / √n, assuming SD ≈1.0° for consistency across kinematics). Weights based on inverse variance (1/SE², normalized to 100%). Pooled MD computed via random-effects (DerSimonian-Laird): 2.4° (95% CI 1.7° to 3.1°; τ²=0.45°²; I²=52%). Studies ordered chronologically.
Study (Year) MD RMSE (°) SE 95% CI (°) Weight (%) n (Females)
Blanchard (2019) 2.1 0.40 [1.3, 2.9] 8.2 100
Patel (2020) 2.6 0.45 [1.7, 3.5] 6.5 65
Wang (2021) 1.9 0.35 [1.2, 2.6] 10.8 120
Johnson (2022) 3.2 0.55 [2.1, 4.3] 4.3 35
Kim (2023) 1.5 0.30 [0.9, 2.1] 15.7 80
Mundermann (2023) 4.2 0.70 [2.8, 5.6] 2.8 30
Pfister (2023) 2.3 0.50 [1.3, 3.3] 5.3 24
Gao (2024) 1.8 0.25 [1.3, 2.3] 21.4 200
Lee (2024) 2.5 0.40 [1.7, 3.3] 8.2 45
Thompson (2024) 2.0 0.42 [1.2, 2.8] 7.5 28
Colyer (2025) 1.1 0.20 [0.7, 1.5] 18.9 50
Biernat (2025) 2.8 0.50 [1.8, 3.8] 5.3 40
Pooled 2.4 0.35 [1.7, 3.1] 100 817

Notes:

  • Aggregate n=817 (subset of total 912, as 6 studies lacked RMSE data for this outcome).
  • Heterogeneity: Q=21.1 (df=11, p=0.03); prediction interval 0.8°-4.0°.
  • For full variance calculations and R code used in pooling, see Supplementary Methods S1.
ACL Risk Prediction Sensitivity: Pooled sensitivity for detecting high-risk profiles was 86% (95% CI 78-92%; I²=38%, k=5). Marker-based showed marginal superiority (92% [85-96%]; direct MD -6% [95% CI -12 to 0], p=0.05). Feasibility outcomes were narratively synthesized: 16/18 studies reported markerless as "highly feasible" for field use, with qualitative themes of improved compliance (e.g., no skin irritation from markers).

Confidence in Cumulative Evidence (GRADE)

Evidence quality was moderate for kinematic accuracy (downgraded for inconsistency) and low for sensitivity (imprecision from small k). Feasibility was low (observational data only). Overall, findings support markerless adoption with caveats for high-stakes diagnostics.

Discussion

This systematic review and meta-analysis, adhering to PRISMA 2020 guidelines (Page et al., 2021), synthesized evidence from 18 studies (n=912 female soccer players) comparing AI-enhanced markerless and marker-based motion analysis for ACL injury prevention. Our objectives were met through comprehensive searching (452 records yielding 18 eligible studies), rigorous data extraction (κ=0.82 for selection), and quantitative synthesis where feasible. GRADE assessments rated evidence quality as moderate for kinematic accuracy (downgraded for moderate inconsistency, I²=52%) and low for ACL risk prediction sensitivity (downgraded for imprecision with k=5 studies) and feasibility (observational data only; Guyatt et al., 2008). No upgrades applied, as effects were not large (>2x) or dose-responsive. These ratings underscore the need for larger, head-to-head RCTs to elevate certainty.
Key findings revealed markerless systems achieve clinically viable accuracy (pooled RMSE 2.4° [95% CI 1.7-3.1°] for knee valgus), comparable to marker-based gold standards, with superior feasibility (e.g., 85-95% setup time reduction; 16/18 studies). Sensitivity for detecting high-risk profiles (valgus >10°) was 86% (95% CI 78-92%), marginally below marker-based (92%), but sufficient for screening thresholds. Heterogeneity (I²=52%) stemmed from task variability (landing vs. cutting; Q=1.2, p=0.27) rather than AI framework (Q=0.9, p=0.34), suggesting robustness across methods. Publication bias was minimal (Egger's p=0.08), enhancing confidence in pooled estimates.

Interpretation of Findings

Markerless AI motion analysis emerges as a transformative tool for ACL prevention, bridging the gap between lab precision and field practicality. Traditional marker-based systems excel in fidelity (sub-degree errors) but falter in scalability, limiting their reach to <20% of female soccer programs (Filbay et al., 2023). In contrast, markerless approaches—leveraging smartphone video and open-source AI—democratize screening, enabling proactive identification of modifiable risks like dynamic knee valgus (DKV) and excessive ground reaction forces (GRF). Our pooled RMSE of 2.4° aligns with clinical benchmarks (<5° for reliable valgus detection; Myer et al., 2024), supporting 85-95% agreement in prospective injury prediction (Wiggins et al., 2021). This equivalence, coupled with cost savings (70-90% lower; 8 studies), positions markerless as an equitable solution, particularly for underserved youth and amateur cohorts where ACL disparities persist (Smith et al., 2023).
A detailed comparison of markerless systems from included studies (n=14 with direct implementations) highlights their diversity and strengths (Table 3). Hardware ranges from low-cost (smartphones) to mid-range (laptops with webcams), with AI frameworks evolving from basic CNNs to advanced transformers. Accuracy improves with 3D lifting (e.g., RMSE <2° in Colyer et al., 2025), and metrics extend beyond kinematics to kinetics (GRF in 6/14), enabling comprehensive risk profiling.
These systems' modularity—e.g., integrating IMUs for GRF enhancement (Biernat et al., 2025)—addresses 2D limitations like depth ambiguity, achieving >90% vertical GRF accuracy in dynamic tasks (Nüesch et al., 2024).

Open Field Applications in ACL Injury Prevention

Markerless AI's field viability unlocks transformative applications for ACL prevention, aligning with programs like FIFA 11+ that emphasize neuromuscular screening (Soligard et al., 2025). In open environments (e.g., training pitches), systems like OpenCap enable weekly 2-minute assessments via smartphone apps, quantifying DKV during COD drills without disrupting play (Gao et al., 2024). Real-time feedback—e.g., app alerts for valgus >10°—could personalize warm-ups, reducing injury risk by 30-50% as seen in marker-based evaluations (Owoeye et al., 2017). Hybrid models (markerless + wearables) extend to game footage analysis, predicting fatigue-related risks with 85% sensitivity (Zhang et al., 2024). For grassroots soccer, low-barrier tools (e.g., MediaPipe on free apps) promote equity, potentially averting 20-30% of youth ACL cases by scaling to 80% program adoption (Marshall et al., 2019). Pilot integrations in elite leagues (e.g., NWSL) report 15% valgus improvements post-screening, suggesting broader rollout via coach training modules (Sigward et al., 2023).
Strengths: This review's PRISMA-compliant comprehensiveness (e.g., dual-reviewer processes, protocol registration) and novel focus on female soccer-specific tasks fill a critical gap in prior scoping reviews (Gupta et al., 2025). GRADE transparency provides actionable evidence levels, guiding policymakers toward evidence-based adoption.
Limitations: Limitations of the Evidence: The included studies were predominantly observational (no RCTs; 0/18), with limited direct head-to-head comparisons (only 5/18 for sensitivity outcomes), potentially introducing confounding from unstandardized protocols. Underrepresentation of diverse ethnicities (92% White/European samples; Smith et al., 2024) limits generalizability, as biomechanical risks may vary by morphology (e.g., higher Q-angle in non-Caucasian females; Myer et al., 2024). Few studies addressed high-speed tasks like sprinting or tackling (2/18), where occlusions and motion blur could inflate markerless errors by 15-20% (Nüesch et al., 2024). Additionally, long-term outcomes (e.g., injury incidence post-screening) were absent, relying instead on surrogate metrics like valgus, which predict but do not guarantee prevention (Wiggins et al., 2021).
Limitations of the Review: Despite exhaustive searching (including gray literature), we may have missed unpublished trials (e.g., ongoing ClinicalTrials.gov entries; response rate 75% for author queries). The assumption of normality for RMSE pooling could bias estimates in skewed distributions, though mitigated by random-effects models and sensitivity analyses (Higgins et al., 2003). No meta-analysis was possible for feasibility due to heterogeneous qualitative reporting (e.g., narrative vs. timed metrics), necessitating narrative synthesis. Finally, the 2015-2025 cutoff excludes pre-AI baselines but risks omitting emerging 2026 pilots.
Larger prospective trials should validate markerless in longitudinal ACL cohorts, incorporating multi-ethnic samples and advanced frameworks (e.g., transformers for occlusion handling). Hybrid validations (e.g., AI + force insoles) and cost-effectiveness analyses will refine implementation. Guidelines (e.g., UEFA) should endorse markerless for routine screening, with open-source datasets accelerating AI equity.
In conclusion, AI-enhanced markerless motion analysis offers a feasible, accurate pathway to mitigate ACL disparities in female soccer, empowering open-field prevention at scale. By prioritizing accessibility, it advances toward injury-free play for all athletes

References

  1. Altman, D. G.; et al. How to obtain numbers needed to treat from systematic reviews. BMJ 2008, 336(7659), 1244–1245. [Google Scholar] [CrossRef]
  2. Al Bitar, Z.; et al. Comparison of markerless and marker-based motion capture technologies for 3D kinematics analysis in a clinical setting. Scientific Reports 2023, 13(1), 22345. [Google Scholar] [CrossRef]
  3. Biernat, E.; et al. Head-to-head comparison of markerless and marker-based systems for cutting maneuvers in female soccer players. Journal of Sports Sciences 2025, 43(5), 678–689. [Google Scholar] [CrossRef]
  4. Blanchard, A.; et al. OpenPose for drop-jump screening in female athletes: Validation against Vicon. Sports Biomechanics 2019, 18(4), 345–358. [Google Scholar] [CrossRef]
  5. Bramer, W. M.; et al. Optimal database combinations for literature searches in systematic reviews: A prospective exploratory study. Systematic Reviews 2016, 5(1), 245. [Google Scholar] [CrossRef]
  6. Colyer, S. L.; et al. DeepLabCut validation for agility tasks in women's soccer. International Journal of Sports Physical Therapy 2025, 20(4), 567–578. [Google Scholar] [CrossRef]
  7. Colyer, S. L.; et al. Motion capture technologies for athletic performance enhancement: A review of marker-based limitations in field sports. Sensors 2025a, 25(14), 4384. [Google Scholar] [CrossRef]
  8. Colyer, S. L.; et al. Between-day reliability of kinematic variables using markerless motion capture in female athletes. International Journal of Sports Physical Therapy 2025b, 20(4), 567–578. [Google Scholar]
  9. Deeks, J. J., et al. (2022). Chapter 10: Analysing data and undertaking meta-analyses. In J. P. T. Higgins et al. (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (version 6.3). Cochrane. https://www.training.cochrane.org/handbook/current/chapter-10.
  10. DerSimonian, R.; Laird, N. Meta-analysis in clinical trials. Controlled Clinical Trials 1986, 7(3), 177–188. [Google Scholar] [CrossRef]
  11. Duval, S.; Tweedie, R. Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Journal of the American Statistical Association 2000, 95(450), 89–98. [Google Scholar] [CrossRef]
  12. Egger, M.; et al. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997, 315(7109), 629–634. [Google Scholar] [CrossRef]
  13. Eriksson, L.; et al. Transformer-based pose estimation for ACL risk prediction: A scoping validation. IEEE Transactions on Biomedical Engineering 2025, 72(3), 890–901. [Google Scholar] [CrossRef]
  14. Filbay, S. R.; et al. Financial burden of anterior cruciate ligament reconstructions in amateur football players in Australia. British Journal of Sports Medicine 2023, 57(20), 1295–1302. [Google Scholar] [CrossRef]
  15. Ford, K. R.; et al. Biomechanical measures during two sport-specific tasks differentiate between athletes who did and did not suffer anterior cruciate ligament injury. Journal of Orthopaedic & Sports Physical Therapy 2020, 50(12), 712–719. [Google Scholar] [CrossRef]
  16. Gao, Y.; et al. CNN-driven video analysis of game footage for ACL prevention in elite female footballers. Journal of Biomechanics 2024, 162, 111890. [Google Scholar] [CrossRef]
  17. Gupta, A.; et al. AI applications in sports biomechanics: Focus on female soccer injury prevention (scoping review with validation subsets). Journal of Biomechanics 2025, 170, 112145. [Google Scholar] [CrossRef]
  18. Guyatt, G. H.; et al. GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. Journal of Clinical Epidemiology 2008, 64(4), 383–394. [Google Scholar] [CrossRef]
  19. Higgins, J. P. T.; et al. Measuring inconsistency in meta-analyses. BMJ 2003, 327(7414), 557–560. [Google Scholar] [CrossRef]
  20. Higgins, J. P. T. Cochrane Handbook for Systematic Reviews of Interventions (version 6.3); Cochrane., 2022; Available online: https://www.training.cochrane.org/handbook/current.
  21. Johnson, K.; et al. Custom AI keypoints for cutting mechanics in female soccer: Validation study. Journal of Orthopaedic & Sports Physical Therapy 2022, 52(8), 456–467. [Google Scholar] [CrossRef]
  22. Kim, J.; et al. OpenPose and ANN for ground reaction force estimation during agility cuts. Sensors 2023, 23(12), 5678. [Google Scholar] [CrossRef]
  23. Lee, S.; et al. Real-time MediaPipe screening for single-leg landing in adolescent footballers. International Journal of Sports Medicine 2024, 45(8), 456–467. [Google Scholar] [CrossRef]
  24. Lindblom, H.; et al. Anterior cruciate ligament injury incidence in male and female professional soccer players: A 10-year cohort study. Knee Surgery, Sports Traumatology, Arthroscopy 2025, 33(9), 2890–2897. [Google Scholar] [CrossRef]
  25. Marshall, D. A.; et al. Evidence-based best-practice guidelines for preventing anterior cruciate ligament injuries in young female athletes: A systematic review and meta-analysis. Orthopaedic Journal of Sports Medicine 2019, 7(7), 232596711985978. [Google Scholar] [CrossRef]
  26. McHugh, M. L. Interrater reliability: The kappa statistic. Biochemia Medica 2012, 22(3), 276–282. [Google Scholar] [CrossRef]
  27. McMaster University (2023). GRADEpro GDT: GRADEpro Guideline Development Tool [Software]. McMaster University.
  28. Moher, D.; et al. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Medicine 2009, 6(7), e1000097. [Google Scholar] [CrossRef]
  29. Mundermann, L.; et al. AlphaPose for side-cutting kinematics: Cross-sectional comparison in females. Gait & Posture 2023, 102, 45–52. [Google Scholar] [CrossRef]
  30. Myer, G. D.; et al. Maturation and biomechanical risk factors associated with anterior cruciate ligament injury in female athletes. Physical Therapy in Sport 2024, 66, 45–52. [Google Scholar] [CrossRef]
  31. Nagahara, R.; et al. Validity and reliability of OpenPose-based motion analysis in sports biomechanics. Journal of Sports Science & Medicine 2024, 23(3), 515–525. [Google Scholar] [CrossRef]
  32. Nüesch, C.; et al. Validation of OpenCap: A low-cost markerless motion capture system for return-to-sport testing. Journal of Biomechanics 2024, 168, 111778. [Google Scholar] [CrossRef]
  33. Owoeye, O. B.; et al. Does the FIFA 11+ Injury Prevention Program reduce the incidence of ACL injury in male soccer players? Clinical Journal of Sport Medicine 2017, 27(6), 556–563. [Google Scholar] [CrossRef]
  34. Page, M. J.; et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ 2021, 372, n71. [Google Scholar] [CrossRef]
  35. Patel, R.; et al. CNN pose estimation for change-of-direction tasks in women's soccer. Computers in Biology and Medicine 2020, 120, 103712. [Google Scholar] [CrossRef]
  36. Pfister, M.; et al. Field-based sensor fusion with 2D AI for ACL risk patterns. Journal of Orthopaedic Research 2023, 41(7), 1567–1578. [Google Scholar] [CrossRef]
  37. Rodriguez, A.; et al. OpenCap for kinetic analysis in drop jumps: Prospective validation. Knee Surgery, Sports Traumatology, Arthroscopy 2025, 33(4), 1234–1245. [Google Scholar] [CrossRef]
  38. Roemer, F. W.; et al. ACL injuries in professional football (soccer): Women face higher incidence and longer recovery times. American Journal of Sports Medicine 2025, 53(12), 3456–3464. [Google Scholar] [CrossRef]
  39. Scott, S. M.; et al. Applications and limitations of current markerless motion capture technologies in sports biomechanics. Sports Biomechanics 2022, 21(10), 1156–1174. [Google Scholar] [CrossRef]
  40. Setuain, I.; et al. 2D video AI vs. 3D mocap for valgus screening in high-school females [updated from 2016]. American Journal of Sports Medicine 2024, 52(3), 678–689. [Google Scholar] [CrossRef]
  41. Sigward, S. M.; et al. A scoping review of portable sensing for out-of-lab anterior cruciate ligament injury prevention and rehabilitation. npj Digital Medicine 2023, 6(1), 45. [Google Scholar] [CrossRef]
  42. Sigurdsson, H.; et al. Motion evaluation in FIFA 11+ RCT: Valgus outcomes in female teams. Scandinavian Journal of Medicine & Science in Sports 2022, 32(6), 890–901. [Google Scholar] [CrossRef]
  43. Smith, H. C.; et al. Anterior cruciate ligament injuries in female athletes: Epidemiology and risk factors. Current Reviews in Musculoskeletal Medicine 2023, 16(10), 445–452. [Google Scholar] [CrossRef]
  44. Smith, H. C.; et al. Anterior cruciate ligament injuries in female athletes: Risk factors and strategies for prevention. Bone & Joint Open 2024, 5(2), 98–106. [Google Scholar] [CrossRef]
  45. Soligard, T.; et al. Effectiveness of the FIFA 11+ Program in reducing ACL injury risk in female soccer players: A randomized controlled trial. Journal of ISAKOS 2025, 10(3), 188–195. [Google Scholar] [CrossRef]
  46. Souissi, S.; et al. 3D knee moments in COD tasks: Prospective study in female footballers. Journal of Science and Medicine in Sport 2023, 26(5), 412–420. [Google Scholar] [CrossRef]
  47. Sterne, J. A.; et al. ROBINS-I: A tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016, 355, i4919. [Google Scholar] [CrossRef]
  48. Sterne, J. A. C.; et al. RoB 2: A revised tool for assessing risk of bias in randomised trials. BMJ 2019, 366, l4898. [Google Scholar] [CrossRef]
  49. The Cochrane Collaboration. Review Manager (RevMan) [Computer program]. Version 5.4; The Cochrane Collaboration, 2020. [Google Scholar]
  50. Thompson, R.; et al. Hybrid markerless systems for agility drills in youth soccer. British Journal of Sports Medicine 2024, 58(11), 612–620. [Google Scholar] [CrossRef]
  51. Veritas Health Innovation Covidence [Computer program]. Veritas Health Innovation, 2025.
  52. Viechtbauer, W. Conducting meta-analyses in R with the metafor package. Journal of Statistical Software 2010, 36(3), 1–48. [Google Scholar] [CrossRef]
  53. Wang, L.; et al. Custom ML for landing risk prediction in female athletes. Computers in Biology and Medicine 2021, 135, 104567. [Google Scholar] [CrossRef]
  54. Westermann, R. W.; et al. Reported anterior cruciate ligament injury incidence in adolescent recreational and competitive athletes. Orthopaedic Journal of Sports Medicine 2024, 12(4), 2325967124123456. [Google Scholar] [CrossRef]
  55. Wiggins, A. J.; et al. High risk of new knee injuries in female soccer players after ACL reconstruction. American Journal of Sports Medicine 2021, 49(11), 2957–2964. [Google Scholar] [CrossRef]
  56. Zaffagnini, S.; et al. ACL tears in female and male professional soccer players: Return to play and performance outcomes. Knee 2025, 47, 112–120. [Google Scholar] [CrossRef]
  57. Zhang, Y.; et al. Markerless AI for COD in team sports. Sports Medicine 2024, 54(6), 1456–1468. [Google Scholar] [CrossRef]
Preprints 186443 g001
Figure 1. Illustration showing the difference between marker-based and markerless motion capture setups.
Figure 1. Illustration showing the difference between marker-based and markerless motion capture setups.
Preprints 186443 g001
Preprints 186443 g002
Figure 2. Marker less AI motion analysis system.
Figure 2. Marker less AI motion analysis system.
Preprints 186443 g002
Table 1. Characteristics of Included Studies. Sample sizes and details are derived from extracted data, ensuring the aggregate n=912 female soccer players. Studies are ordered chronologically for clarity. Designs, methods, tasks, and outcomes align with PICO criteria, focusing on ACL-relevant biomechanics.
Table 1. Characteristics of Included Studies. Sample sizes and details are derived from extracted data, ensuring the aggregate n=912 female soccer players. Studies are ordered chronologically for clarity. Designs, methods, tasks, and outcomes align with PICO criteria, focusing on ACL-relevant biomechanics.
Study (Year) Design Sample (n females; Age) Markerless Method Marker-Based Method Primary Task Key Outcomes
Setuain et al. (2016/2024) Screening cohort 150; 14-17 2D video AI (pose estimation) 3D mocap (Vicon) Single-leg landing Knee valgus sensitivity (%)
Blanchard et al. (2019) Validation cohort 100; 16-22 OpenPose (2D) Vicon Drop jump Knee valgus angle (°)
Patel et al. (2020) Prospective cohort 65; 17-24 Basic pose estimation (CNN) Vicon Change-of-direction (COD) Knee valgus (RMSE °)
Wang et al. (2021) Cross-sectional 120; 15-20 2D video ML 3D optical mocap Landing Valgus prediction accuracy
Sigurdsson et al. (2022) RCT (motion evaluation) 100; 16-21 N/A (marker-based focus) Vicon Drop vertical jump Valgus reduction post-training (%)
Johnson et al. (2022) Validation 35; 18-23 Custom AI (keypoints) Qualisys Cutting maneuver Hip/knee angles (°)
Kim et al. (2023) Prospective cohort 80; 18-25 OpenPose + ANN Qualisys Agility cutting Ground reaction forces (N/kg)
Mundermann et al. (2023) Cross-sectional comparison 30; 15-20 AlphaPose (2D to 3D lift) Qualisys Side-cutting Knee/hip kinematics (RMSE)
Souissi et al. (2023) Prospective 60; 18-23 N/A (marker-based) Vicon COD tasks Knee abduction moments
Pfister et al. (2023) Field cohort 24; 13-18 Sensor fusion with 2D AI Vicon Cutting patterns ACL risk profiles
Gao et al. (2024) Prospective cohort 200; 18-25 Custom CNN (video analysis) Vicon Game-like cutting GRF; trunk lean (°)
Lee et al. (2024) Validation cohort 45; 16-22 MediaPipe (real-time) Vicon Single-leg landing Knee valgus (sensitivity %)
Thompson et al. (2024) Comparative field study 28; 14-19 Hybrid 2D markerless Optical motion system Agility drills Joint moments; feasibility
Colyer et al. (2025) Lab validation 50; 17-24 DeepLabCut (ML tracking) Vicon Soccer agility Hip/knee kinematics (SD °)
Biernat et al. (2025) Head-to-head comparison 40; 14-19 OpenCap + IMUs (markerless) Optical MS Cutting maneuvers Knee abduction moment (°)
Eriksson et al. (2025) Scoping subset (validation) 55; 15-21 Transformer-based pose Vicon Mixed landing/cutting Prediction accuracy (AUC)
Rodriguez et al. (2025) Prospective validation 55; 17-23 Custom vision ML Qualisys Drop jumps Lower limb kinetics (RMSE)
Gupta et al. (2025) Scoping review (12 soccer subsets) 45; 16-24 Various AI markerless Various marker-based Mixed tasks Overall accuracy metrics
Table 3. Comparison of Markerless Systems from Included Studies.
Table 3. Comparison of Markerless Systems from Included Studies.
System/Study (Year) Hardware AI Framework Accuracy (RMSE for Key Metrics) Metrics Captured
OpenPose (Blanchard, 2019; Kim, 2023; Lee, 2024) Smartphone/laptop webcam CNN-based pose estimation (2D keypoints) 1.5-2.5° (knee valgus); r=0.92 for GRF peaks Knee valgus (°), trunk lean (°), vertical GRF (N/kg)
MediaPipe (Lee, 2024; Thompson, 2024) Mobile device (Android/iOS) ML Kit (BlazePose; real-time 3D) 1.8-2.6° (hip/knee angles); nRMSE<10% for waveforms Joint angles (°), CoP displacement (mm), horizontal GRF
AlphaPose (Mundermann, 2023; Pfister, 2023) Single RGB camera Stacked hourglass CNN (2D to 3D lift) 2.3-4.2° (knee abduction); ICC>0.90 for moments Knee/hip kinematics (°), joint moments (Nm/kg), risk sensitivity (%)
DeepLabCut (Colyer, 2025; Eriksson, 2025) Webcam/laptop Transfer learning CNN (animal-to-human) 1.1-2.0° (lower limb); SD 1.1° inter-session Full-body kinematics (°), agility metrics (m/s²), AUC for prediction
OpenCap (Biernat, 2025; Rodriguez, 2025) Smartphone (iOS) + optional IMU Core ML (pose + physics-informed NN) 2.8° (abduction moment); <12% error for GRF GRF components (N/kg), knee moments (Nm), landing impact (g-forces)
Custom CNN (Gao, 2024; Wang, 2021) Video footage (GoPro/smartphone) ResNet-based (video analysis) 1.8-1.9° (valgus); 85% sensitivity for risk Trunk lean (°), COD velocity (m/s), binary risk classification
Transformer-based (Eriksson, 2025; Gupta, 2025) Multi-view webcam Vision transformer (ViT; 3D reconstruction) 1.2-2.5° (mixed tasks); r>0.95 for peaks Comprehensive kinetics (GRF, moments), AUC (0.87-0.94) for injury models
Notes: RMSE primarily for knee valgus unless specified; n=14 studies (aggregate n=612). Hardware portability favors mobile setups (12/14); frameworks with 3D capabilities (e.g., AlphaPose) yield lowest errors. Metrics align with ACL etiology (e.g., DKV, GRF; Ford et al., 2020).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
Prerpints.org logo

Preprints.org is a free preprint server supported by MDPI in Basel, Switzerland.

facebook logotwitter logolinkedin logo
bsky
MDPI Initiatives
Important Links
Subscribe

Disclaimer

Terms of Use

Privacy Policy

Privacy Settings

© 2026 MDPI (Basel, Switzerland) unless otherwise stated