Document is current

Article History

Received: 10 October 2025

Accepted: 2 June 2026

First Online: 14 June 2026

Declarations

:

: The authors declare no competing interests.

: This study was conducted in accordance with the principles of the Declaration of Helsinki. The work was designed and carried out as a low-risk, non-interventional engineering and technology feasibility evaluation involving only non-invasive wearable sensor monitoring of community-dwelling older adults, with no medical intervention, no diagnostic or therapeutic claims, and no use of investigational drugs or medical devices. Under the institutional research governance policies of HuaiYin Institute of Technology and Nagasaki Institute of Applied Science that apply to non-interventional technology feasibility studies of this kind, formal ethics committee review was not required, and therefore no ethics approval number was issued for this work. The institutional research administration offices of both institutions reviewed the protocol prior to participant enrollment and confirmed that the study fell outside the scope of mandatory Institutional Review Board oversight. Notwithstanding the absence of a requirement for formal ethics approval, all participants provided written informed consent prior to enrollment in the study, in line with good research practice for studies involving human participants. The research protocol included comprehensive participant information regarding the purpose, procedures, potential risks, benefits, and data handling practices. Participants were informed of their right to withdraw from the study at any time without penalty. Special consideration was given to the vulnerable nature of the elderly population, ensuring that consent procedures were appropriate for participants’ cognitive abilities and that family members or legal guardians were involved where necessary.

: The primary objective of this study was to evaluate the technical performance of a multimodal sensor fusion system and its deep learning prediction pipeline under real-world operating conditions—specifically, classification accuracy, inference latency, false alarm rate, and system usability. The three-month deployment phase involving 40 intervention and 40 control participants was designed as an engineering field test to assess whether the system could function reliably over an extended period in naturalistic settings, not as a trial intended to establish the clinical efficacy of a therapeutic or preventive intervention. The wearable unit is a research prototype that has not been submitted for medical device certification, and no diagnostic or therapeutic claims were made to participants during the study. Fall incidence data were recorded as part of routine safety monitoring and are reported here as an exploratory, secondary outcome to contextualize system performance; the study was neither designed nor statistically powered to demonstrate a causal intervention effect on fall rates. Under the ICMJE framework, purely observational studies and technology feasibility evaluations in which the health-related outcome is not the primary endpoint driving group assignment do not mandate prospective trial registration. The institutional research administration offices of both participating institutions reviewed this classification at the time of protocol design and confirmed that the engineering field-validation nature of the study did not require prospective clinical trial registration. Nonetheless, we acknowledge that the between-group comparison of fall incidence could be interpreted as resembling an interventional design. To maintain full transparency, we have clearly labeled the fall reduction finding as exploratory and associational throughout the manuscript, and we strongly recommend that any future study seeking to establish the clinical efficacy of the warning system be prospectively registered as a randomized controlled trial before enrollment begins.