EUVIMEDEUVIMED entdecken
Systematische ÜbersichtMetaanalyseOpen Access

Machine Learning and Deep Learning Models for Predicting Future Falls in Community-Dwelling Older Adults: Systematic Review and Meta-Analysis of Longitudinal Evidence

Machine-Learning- und Deep-Learning-Modelle zur Vorhersage künftiger Stürze bei älteren Menschen in der Gemeinschaft: systematische Übersicht und Metaanalyse longitudinaler Evidenz

Ying Gao, Doudou Xu, Xinru Li, Jue Wang, Linbin Wang, Beiwen Wu, Haifeng Zhao, Xian Qiu, Weiyi Zhu

Journal of Medical Internet Research · 14.05.2026 · 28 · e84844-e84844

DOI: 10.2196/84844 ↗

Auf einen Blick

28 Studien zeigen für ML-/DL-Modelle zur Vorhersage künftiger Stürze bei älteren Menschen in der Gemeinschaft eine gepoolte AUC von 0,79. Die Streuung war sehr hoch, die externe Validierung selten.

Fragestellung

Wie gut sagen ML-/DL-Modelle künftige Stürze bei älteren, in der Gemeinschaft lebenden Menschen voraus?

Methode

Systematische Übersicht und Metaanalyse longitudinaler Studien; Suche in 6 Datenbanken bis 23. September 2024 mit Updates am 31. August 2025 und 28. Februar 2026; PROBAST-Bewertung; AUC-Metaanalyse mit Hartung-Knapp-Sidik-Jonkman.

Ergebnisse

28 Studien nach Screening von 10.253 Treffern. Gepoolte AUC 0,79 (95 %-KI 0,69–0,87), 95 %-Prädiktionsintervall 0,20–0,99; Heterogenität sehr hoch (I2=99,8 %). Nur 1 Modell extern validiert.

Einschränkungen

Alle Modelle mit hohem Verzerrungsrisiko; Kalibrierung selten berichtet; breite Prädiktionsintervalle und geringe externe Validierung.

Schlussfolgerung der Autor:innen

ML-/DL-Modelle zeigen Potenzial zur Identifikation älterer Menschen mit erhöhtem zukünftigen Sturzrisiko; die reale Leistung könnte wegen breiter Prädiktionsintervalle und hohem Verzerrungsrisiko zu optimistisch sein.

Kernaussagen

  • Gepoolte AUC: 0,79 (95 %-KI 0,69–0,87)
  • 95 %-Prädiktionsintervall: 0,20–0,99
  • Nur 1 Modell extern validiert; alle Modelle hohes Bias-Risiko

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Systematische Übersicht in der Zeitschrift Journal of Medical Internet Research, online erschienen am 14.05.2026. Verfasst von Ying Gao und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Nursing, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, No.197, 2nd Ruijin Rd, Huangpu District, Shanghai, 200025, …. Hauptthemen laut MEDLINE: Accidental Falls, Deep Learning, Machine Learning, Independent Living. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 108 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

Auf einen BlickAbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Machine learning (ML) and deep learning (DL) show promise for fall risk prediction, but prior reviews focused mainly on real-time fall detection, in-hospital falls, or conventional statistical models. The performance of ML-DL–based models for predicting future falls in community-dwelling older adults remains unclear.

Objective

This study aimed to review ML-DL studies for predicting future falls among community-dwelling older adults and meta-analyze discrimination where feasible.

Methods

Six databases were searched from inception to September 23, 2024, with updates on August 31, 2025, and February 28, 2026. We included longitudinal studies developing or validating ML-DL models to predict future falls in community-dwelling adults aged ≥60 years and excluded real-time detection, simulated or no fall, and inpatient studies. Risk of bias was assessed using the Prediction Model Risk of Bias Assessment Tool (PROBAST). Areas under the curve (AUCs) were meta-analyzed using Hartung-Knapp-Sidik-Jonkman random-effects models with 95% CIs. Heterogeneity, 95% prediction intervals (PIs), sensitivity analyses, and subgroup analyses were conducted.

Results

After screening 10,253 records, 28 (0.3%) studies were included; 18 (64.3%) focused on general older adults. Prediction horizons ranged from 3 months to 7 years, and fall incidence ranged from 1.6% to 46.6%. Twenty-three (82.1%) studies applied ML, and 5 (17.9%) studies used DL. Input modalities included text (n=18, 64.3%), sensor (n=5, 17.9%), image (n=1, 3.6%), and multimodal data (n=4, 14.3%). Common predictors included age, sex, fall history, depression, and basic daily activities. Only one model underwent external validation. Calibration reporting was sparse. All models were rated at high risk of bias. Ten models were meta-analyzed, yielding a pooled AUC of 0.79 (95% CI 0.69‐0.87) with extreme heterogeneity ( τ 2 =0.64; τ =0.80; I 2 =99.8%; Q =4128.99). The confidence-distribution bootstrap PI was 0.20 to 0.99, indicating substantial uncertainty in expected performance across new populations. Subgroup analyses indicated moderation by sample size and population type, with higher discrimination in specific populations than in general samples; however, the specific population subgroup included only 2 studies. Although all participants were community dwelling, some cohorts were recruited through clinically enriched pathways rather than general community sampling.

Conclusions

ML-DL models show potential for identifying community-dwelling older adults at elevated future fall risk; however, wide PIs, limited external validation, and high risk of bias suggest real-world performance may be optimistic. The pooled AUC should be interpreted as a summary of reported discrimination under study-specific conditions, predominantly from internally validated, high-risk-of-bias models, rather than as a robust estimate of transportable real-world performance. This review extends prior reviews by focusing on community-dwelling settings and by integrating PROBAST, Hartung-Knapp-Sidik-Jonkman meta-analysis, PIs, and modality-specific synthesis to evaluate both discrimination and uncertainty. Findings support the use of ML-DL models for proactive fall prevention while emphasizing the need for validation and context-specific implementation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Accidental FallsDeep LearningMachine LearningIndependent LivingHumansAgedLongitudinal StudiesAging in PlacePredictive Learning ModelsPrediction Algorithms

Stichwörter der Autor:innen

community-dwelling older adultsfallsmachine learningmeta-analysisrisk predictionsystematic review

Autor:innen und Institutionen

  • Department of Nursing, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, No.197, 2nd Ruijin Rd, Huangpu District, Shanghai, 200025, China, 86 02164370045.
  • School of Nursing, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
  • Department of Orthopedics, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
  • Department of Neurosurgery, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China.
  • Department of Psychiatry, University of Cambridge, Cambridge, United Kingdom.
  • Institute of Medical Artificial Intelligence, Shanghai Jiao Tong University School of Medicine, Shanghai, China.

Zitieren

APA

Gao, Y., Xu, D., Li, X., Wang, J., Wang, L., Wu, B., Zhao, H., Qiu, X., & Zhu, W. (2026). Machine Learning and Deep Learning Models for Predicting Future Falls in Community-Dwelling Older Adults: Systematic Review and Meta-Analysis of Longitudinal Evidence. Journal of Medical Internet Research, 28, e84844-e84844. https://doi.org/10.2196/84844

Vancouver

Gao Y, Xu D, Li X, Wang J, Wang L, Wu B, et al. Machine Learning and Deep Learning Models for Predicting Future Falls in Community-Dwelling Older Adults: Systematic Review and Meta-Analysis of Longitudinal Evidence. Journal of Medical Internet Research. 2026;28:e84844-e84844. doi:10.2196/84844

BibTeX
@article{Gao2026,
  author = {Gao, Ying and Xu, Doudou and Li, Xinru and Wang, Jue and Wang, Linbin and Wu, Beiwen and Zhao, Haifeng and Qiu, Xian and Zhu, Weiyi},
  title = {Machine Learning and Deep Learning Models for Predicting Future Falls in Community-Dwelling Older Adults: Systematic Review and Meta-Analysis of Longitudinal Evidence},
  journal = {Journal of Medical Internet Research},
  year = {2026},
  volume = {28},
  pages = {e84844--e84844},
  doi = {10.2196/84844}
}
Kostenlos starten →