{"resourceId":"nature-ma-shap-university-football-injury-2025-11-17","versions":[{"version":"external-36f9879e249b6709aac1ab43a1c1dd82fd26e1516babbde3a77af4cb2cd4faea","resource":{"id":"nature-ma-shap-university-football-injury-2025-11-17","title":"Scientific Reports: university football injury ML hits high metrics but needs prospective validation","organization":"Dankook University; Southwest Minzu University; collaborating authors","sector":"University football / sports medicine ML research","geography":"China (Kaggle university football dataset; transfer limits to U.S. NCAA)","publishedAt":"17 November 2025","publicationDate":"2025-11-17","eventDate":null,"sourceName":"Scientific Reports (Nature Portfolio)","sourceLabel":"Peer-reviewed academic research","sourceUrl":"https://www.nature.com/articles/s41598-025-24144-y","evidenceClass":"academic-research","outcomeClass":"cautionary","topics":["data-security","governance-procurement","operating-model","accessibility-workforce"],"finding":"Ma, Liu, and Pei (published 2025-11-17) train interpretable ML injury-risk models on a Kaggle dataset of 800 Chinese university football players (18 features). SVM reached 95.6% accuracy / 99.2% ROC-AUC; SHAP ranked stress, sleep, and balance above many traditional fitness metrics. Authors explicitly warn the single-dataset, non-externally validated design limits generalizability and that prospective validation is essential before clinical deployment, while discussing ethical risks of high-risk labels for student-athletes. Why now: counters same-week vendor/coaching AI enthusiasm with university-specific injury-model limits after recent collegiate wearable/injury-AI coverage.","sledRelevance":"University-athlete injury ML with strong retrospective metrics and equally strong deployment cautions—including stigma/ethics of risk labels—directly relevant to collegiate sports medicine AI procurement.","evidence":"Comparative evaluation of 10 algorithms, SHAP explanations, calibration/Brier discussion, and an extensive limitations section contrasting AUC 0.99 with prior literature (≈0.76–0.83).","architectureImplications":"Baseline tabular ML with explainability layer suitable for decision support—not real-time autonomous load control; dynamic load features were missing and flagged as future work.","governanceImplications":"TRIPOD-style external/prospective validation and ethical deployment (confidentiality, support resources, avoid deterministic labels) are treated as prerequisites to clinical use.","securityPrivacyImplications":"Uses de-identified public dataset; applying similar models on live NCAA athletes would require FERPA-aware handling and careful disclosure of stress/mental-health features.","caveats":"Kaggle/public dataset (not multi-site NCAA EMR); perfectly balanced classes may inflate metrics; China university football context; authors flag possible dataset artifact behind near-perfect AUC.","streamIds":["college-athletics"],"roles":{"sales":"Sports medicine and performance leaders are pitched injury AI with impressive accuracy slides. Discovery should ask for external validation, prospective calibration at real injury base rates, and how high-risk labels are communicated to athletes. Bound value is interpretable screening hypotheses (stress/sleep/balance)—not deployable clinical automation. Unsupported claims include 95%+ real-world accuracy, causal prevention from SHAP rankings, or readiness for NCAA roster decisions.","engineering":"Fit is offline, explainable risk stratification research tooling. Architecture should separate model scoring from coach-facing workflows, keep humans in the loop, and avoid leaking mental-health features broadly. Prerequisites: local labeled data, temporal validation, and ethics review. Constraints: single public dataset and static baselines. Security: protect stress/PSS-type features as sensitive. Proof of value: prospective season holdout with clinically useful sensitivity at realistic prevalence—not retrospective AUC alone.","delivery":"Delivery is a research-to-clinic staged path: replicate on local data, validate prospectively, then pilot decision-support with counseling resources. Dependencies: sports medicine ownership and counseling capacity. Ownership: team physician/ATC for clinical use; data science for model maintenance. Skills adoption teaches staff to interpret SHAP drivers without stigmatizing athletes. Governance checkpoints: IRB/ethics, confidentiality rules, and ban on playing-time automation. Acceptance criteria: external validation report; athlete communication protocol. Risks include false reassurance, anxiety from labels, and overfitting to curated data."},"roleDetails":{"sales":{"customerProblem":"Collegiate sports medicine wants injury-risk tools tailored to university athletes, but vendor metrics often skip external validation and ethics.","stakeholders":["Head athletic trainer / team physician","Sport coaches","Student-athlete wellness / counseling","Compliance / privacy"],"discoveryQuestions":["Has the model been prospectively validated at this school’s injury base rate?","How are high-risk labels shared, and with whom?","Which features (stress, sleep) are treated as sensitive mental-health data?"],"valueHypothesis":"Interpretable models can prioritize modifiable factors like stress and sleep if used as decision support after local validation.","potentialEngagement":"Season-long prospective validation study with predefined clinical utility thresholds before any roster-facing use.","unsupportedClaims":["95.6% real-world accuracy on NCAA teams","SHAP importance proves causal prevention targets","Safe to automate playing-time or scholarship decisions"]},"engineering":{"fit":"Explainable tabular injury-risk research model for university football—not an autonomous clinical device.","architectureIntegration":"Batch scoring from baseline assessments → SHAP explanations → clinician review queue; no direct write-back to practice plans.","prerequisites":["Local longitudinal labels","Temporal train/test discipline","Sensitive-feature access controls"],"constraints":["Public balanced dataset may overstate discrimination","Static baseline cannot capture mid-season load spikes","China-to-NCAA transfer unproven"],"security":"Treat stress and related psychometrics as confidential health-adjacent data; minimize coach-wide exposure.","proofOfValue":["Prospective AUC/calibration at real prevalence","Decision-curve or net-benefit analysis","Athlete outcome monitoring for label-related harm"]},"delivery":{"work":"Replicate features locally, validate prospectively, train clinicians on explanation use, and publish an athlete communication protocol.","dependencies":["Sports medicine data pipeline","Counseling resources for high-stress flags","Ethics/IRB approval"],"ownership":"Team physician/ATC owns clinical interpretation; analytics owns model monitoring.","skillsAdoption":"Teach staff to pair SHAP drivers with supportive interventions rather than punitive labels.","governanceCheckpoints":["Pre-deployment ethics review","Mid-season calibration check","Ban on automated selection decisions"],"acceptanceCriteria":["External/prospective validation memo signed by sports medicine","Written confidentiality and support pathway","Documented false-negative/positive review"],"risks":["Stigma and anxiety from high-risk labels","Overfit metrics driving procurement","Neglect of dynamic load monitoring"]}},"retrievedAt":"2026-09-28T03:19:30Z","enrichedAt":"2026-09-28T03:30:00Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Highlights university athletes’ dual academic-athletic stress as central risk context; implies wellness/counseling capacity must accompany any risk-model rollout.","procurementImplications":"Discount single-dataset accuracy claims; require prospective multi-site validation and ethical deployment plans before purchase.","operatingModelImplications":"Injury AI should remain decision support under sports medicine, with lifestyle/psychosocial interventions prioritized over black-box automation.","sourceVerification":{"openedUrl":"https://www.nature.com/articles/s41598-025-24144-y","referenceExcerpt":"Prospective validation is essential before clinical deployment.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}