EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

DIGITAL PHENOTYPING AND ARTIFICIAL INTELLIGENCE FOR PREDICTING MUSCULOSKELETAL REHABILITATION OUTCOMES

Bohdana Doskaliuk

Anti-Aging Eastern Europe · 30.06.2026 · 5 (2) · 65-74

DOI: 10.56543/aaeeu.2026.5.2.01 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Anti-Aging Eastern Europe, online erschienen am 30.06.2026. Verfasst von Bohdana Doskaliuk. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Rehabilitation outcomes in musculoskeletal disorders vary considerably among patients with similar diagnoses and baseline characteristics. Conventional prognostic assessment relies mainly on measurements obtained during isolated clinical encounters and may therefore overlook fluctuations in symptoms, physical activity, adherence, movement behaviour, and early treatment response. Digital health technologies enable repeated, real-world data collection, although digital monitoring does not automatically generate clinically useful predictions. This narrative review examines the contribution of wearable sensors, mobile applications, electronic patient-reported outcome measures, telerehabilitation platforms, computer vision, and artificial intelligence to predicting musculoskeletal rehabilitation outcomes. A PubMed/MEDLINE and Scopus search was conducted for peer-reviewed English-language publications available up to May 2026. Search concepts included musculoskeletal disorders, rehabilitation, digital health technologies, artificial intelligence, prognosis, treatment response, and functional recovery. Recent reviews, methodological guidance, and representative prognostic studies were prioritised. Wearables, smartphones, electronic questionnaires, remote rehabilitation platforms, and camera-based systems can provide longitudinal data on physical activity, movement quality, symptoms, sleep, engagement, and exercise adherence. Machine-learning methods can integrate these data with clinical, functional, and psychosocial variables to predict pain persistence, non-response, delayed recovery, poor adherence, and additional healthcare use. However, digital exclusion, algorithmic bias, privacy, data ownership, device dependency, and unclear professional accountability remain important barriers. Digital technologies may shift musculoskeletal rehabilitation from episodic assessment towards longitudinal, adaptive, and data-informed care. Their clinical value will depend on reliable digital measures, meaningful outcomes, appropriately validated prediction models, and workflows in which predictions support rather than replace clinician–patient decision-making. Future research should prioritise multimodal longitudinal models, external and prospective validation, transparent reporting, equitable implementation, and trials assessing whether prediction-supported care improves patient-centred outcomes.

Zitieren

APA

Doskaliuk, B. (2026). DIGITAL PHENOTYPING AND ARTIFICIAL INTELLIGENCE FOR PREDICTING MUSCULOSKELETAL REHABILITATION OUTCOMES. Anti-Aging Eastern Europe, 5(2), 65-74. https://doi.org/10.56543/aaeeu.2026.5.2.01

Vancouver

Doskaliuk B. DIGITAL PHENOTYPING AND ARTIFICIAL INTELLIGENCE FOR PREDICTING MUSCULOSKELETAL REHABILITATION OUTCOMES. Anti-Aging Eastern Europe. 2026;5(2):65-74. doi:10.56543/aaeeu.2026.5.2.01

BibTeX
@article{Doskaliuk2026,
  author = {Doskaliuk, Bohdana},
  title = {DIGITAL PHENOTYPING AND ARTIFICIAL INTELLIGENCE FOR PREDICTING MUSCULOSKELETAL REHABILITATION OUTCOMES},
  journal = {Anti-Aging Eastern Europe},
  year = {2026},
  volume = {5},
  number = {2},
  pages = {65--74},
  doi = {10.56543/aaeeu.2026.5.2.01}
}
Kostenlos starten →