Prediction of safety accident subtypes for persons with dementia using sensors and machine learning: an observational study.
Background and Objectives We explored the use of machine learning models for predicting safety accident subtypes among individuals with dementia using in-home sensors and to identify key predictors. Research Design and Methods An observational study was conducted using 966 days of in-home sensor dat...
| Publicado en: | Gerontologist Vol. 66; no. 3; pp. 1 - 11 |
|---|---|
| Autores principales: | , , , , , , |
| Formato: | Artículo |
| Publicado: |
Oxford University Press / USA
Mar2026
|
| Materias: | |
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ssf&AN=192099527&site=ehost-live header: @attributes: shortDbName: ssf uiTerm: 192099527 longDbName: Social Sciences Full Text (H.W. Wilson) uiTag: AN controlInfo: bkinfo: jinfo: jid: 00169013 GET jtl: Gerontologist issn: 00169013 maglogo: N pubinfo: dt: Mar2026 vid: 66 iid: 3 pid: 622 pub: Oxford University Press / USA artinfo: ui: 192099527 10.1093/geront/gnaf313 ppf: 1 ppct: 10 formats: tig: atl: Prediction of safety accident subtypes for persons with dementia using sensors and machine learning: an observational study. aug: au: Yang, Eunjin Lee, Ji Yeon Choi, YeonKyu Lee, SungHee Jang, YoonHyung Cho, Aeyoung Lee, Kyung Hee affil: College of Nursing, Research Institute of AI and Nursing Science, Gachon University, Incheon, Republic of Korea Department of Nursing, Inha University, Incheon, Republic of Korea BRFrame Inc, Seoul, Republic of Korea Mo-Im Kim Nursing Research Institute, Yonsei University College of Nursing, Seoul, Republic of Korea su: South Korea Risk-taking behavior Violence Caregivers Self-mutilation Analysis of variance Injury risk factors Accidents Risk assessment Boosting algorithms Prediction models Crush syndrome Electroconvulsive therapy Burns & scalds Research funding Sleep latency Receiver operating characteristic curves Home safety Wandering behavior Invective Scientific observation Questionnaires Logistic regression analysis Wearable technology Respiratory obstructions Foreign bodies Chi-squared test Descriptive statistics Electronic equipment Sleep duration Classification algorithms Support vector machines Prediction algorithms Diary (Literary form) Neuropsychological tests Research methodology Machine learning Data analysis software Comparative studies Dementia patients Time Algorithms Predictive validity Physical activity Accidental falls Motion capture (Human mechanics) Evaluation sug: subj: Risk-taking behavior Violence Caregivers Self-mutilation Analysis of variance South Korea Residential building construction Architectural Services Other Building Finishing Contractors Semiconductor and other electronic component manufacturing Other Electronic Component Manufacturing Electronic components, navigational and communications equipment and supplies merchant wholesalers Consumer Electronics Repair and Maintenance Injury risk factors Accidents Risk assessment Boosting algorithms Prediction models Crush syndrome Electroconvulsive therapy Burns & scalds Research funding Sleep latency Receiver operating characteristic curves Home safety Wandering behavior Invective Scientific observation Questionnaires Logistic regression analysis Wearable technology Respiratory obstructions Foreign bodies Chi-squared test Descriptive statistics Electronic equipment Sleep duration Classification algorithms Support vector machines Prediction algorithms Diary (Literary form) Neuropsychological tests Research methodology Machine learning Data analysis software Comparative studies Dementia patients Time Algorithms Predictive validity Physical activity Accidental falls Motion capture (Human mechanics) Evaluation keyword: Dementia Safety Sensors Dementia Safety Sensors ab: Background and Objectives We explored the use of machine learning models for predicting safety accident subtypes among individuals with dementia using in-home sensors and to identify key predictors. Research Design and Methods An observational study was conducted using 966 days of in-home sensor data, sleep data from wearable Actiwatch devices, caregiver-completed structured safety accident diary data, and individual data collected in South Korea. Five machine learning classification models were developed to predict physical injury, nighttime behaviors/wandering, and risky behaviors. Model performance was compared, and the most important predictive features were extracted. Results The Gradient Boosting Machine showed the best performance in predicting physical injury and nighttime behaviors, while CatBoost performed best for risky behaviors. Activity patterns recorded using in-home sensors emerged as essential features for predicting different safety accident subgroups, particularly for nighttime behaviors and wandering. Discussion and Implications These findings highlight the potential of these technologies to identify high-risk individuals with dementia. Further research is recommended to integrate these methods for daily safety monitoring of this population. pubtype: Academic Journal doctype: Article src: R language: English refInfo: copyright: @attributes: flag: N holdings: @attributes: islocal: N |
|---|