Construcción de un prototipo de registro de caídas basado en machine learning para mayores institucionalizados.
Saved in:
| Title: | Construcción de un prototipo de registro de caídas basado en machine learning para mayores institucionalizados. |
|---|---|
| Alternate Title: | Development of a fall registration prototype based on machine learning for institutionalized older adults. Construção de um protótipo de registro de quedas baseado em machine learning para idosos institucionalizados. |
| Authors: | Dinamarca-Montecinos, José Luis1 (AUTHOR) jose.dinamarca@uv.cl, Flores-Moraga, María Jesús2 (AUTHOR) jesuflores.382@gmail.com, Durán-Novoa, Roberto Alejandro2 (AUTHOR) durannovoa@uwalumni.com, Briede-Westermeyer, Juan Carlos2 (AUTHOR) jcbriede@gmail.com |
| Source: | MedUNAB. apr-jul2025, Vol. 28 Issue 1, p154-169. 16p. |
| Subjects: | MACHINE learning, LONG-term care facilities, SAFETY, MEDICAL care, PARTICIPATORY design, OLDER people, DATA analysis |
| Geographic Terms: | CHILE |
| Abstract (English): | Introduction. : Falls in institutionalized older adults represent an underestimated public health problem associated with disability, dependence and mortality. In Chile, the absence of standardized records in long-term care facilities (LTCF) for older adults limits effective prevention. The objective of this study was to design a prototype of a digital fall registration system based on machine learning (ML) for its implementation in LTCF. Methodology. : The double diamond design methodology was used in four phases: a) identifying stakeholders and gathering information through qualitative interviews; b) analyzing causes and prioritizing ideas with an Analytical Hierarchy Process (AHP) and Pugh matrices; c) conceptual design and generating the minimum viable product (MVP), and d) developing prototypes for usability validation. Results. : There was evidence of great heterogeneity in the current records and a lack of subsequent data analysis. A MVP was developed, which includes a form for recording falls, a visualization of preventive measures, differentiated user profiles and educational tools. The system was internally validated by caregivers, managers and health care professionals in LTCF. Discussion. : Using ML would make it possible to automate data analysis and customize preventive measures. The participatory design and preventive approach were key to its acceptability. Conclusions. : The developed prototype has the potential to optimize how falls are recorded in LTCF, improve prevention and strengthen care for institutionalized older adults. [ABSTRACT FROM AUTHOR] |
| Abstract (Spanish): | Introducción. : Las caídas en personas mayores institucionalizadas representan un problema de salud pública subestimado, asociado a discapacidad, dependencia y mortalidad. En Chile, la ausencia de registros estandarizados en establecimientos de larga estadía para adultos mayores (ELEAM) limita la prevención efectiva. Este estudio tuvo como objetivo diseñar un prototipo de sistema digital de registro de caídas basado en aprendizaje automático, o machine learning (ML), para su implementación en ELEAM. Metodología. : Se empleó la metodología de diseño de doble diamante en cuatro fases: a) identificación de actores y levantamiento de información mediante entrevistas cualitativas; b) análisis de causas y priorización de ideas con matrices Analytical Hierarchy Process (AHP) y Pugh; c) diseño conceptual y generación del producto mínimo viable (PMV), y d) elaboración de prototipos para validación de usabilidad. Resultados. : Se evidenció una gran heterogeneidad en los registros actuales y ausencia de análisis posterior de datos. Se desarrolló un PMV que incluye un formulario de registro de caídas, visualización de medidas preventivas, perfiles de usuario diferenciados y herramientas educativas. El sistema fue validado internamente por cuidadores, directivos y profesionales de salud en ELEAM. Discusión. : El uso de ML permitiría automatizar el análisis de datos y personalizar medidas preventivas. El diseño participativo y el enfoque preventivo fueron claves para su aceptabilidad. Conclusiones. : El prototipo desarrollado tiene potencial para optimizar el registro de caídas en ELEAM, mejorar la prevención y fortalecer la atención en personas mayores institucionalizadas. [ABSTRACT FROM AUTHOR] |
| Abstract (Portuguese): | Introdução. : Quedas em idosos institucionalizados representam um problema de saúde pública subestimado, associado à deficiência, dependência e mortalidade. No Chile, a falta de registros padronizados em instituições de longa permanência para idosos (ILPI) limita a prevenção eficaz. Este estudo teve como objetivo projetar um protótipo de sistema digital de registro de quedas baseado em aprendizado de máquina, ou machine learning (ML), para sua implementação em ILPI. Metodologia. : Foi utilizada a metodologia de design duplo-diamante em quatro fases: a) identificação de stakeholders e coleta de informações por meio de entrevistas qualitativas; b) análise de causas e priorização de ideias com matrizes Analytical Hierarchy Process (AHP) e Pugh; c) design conceitual e geração do produto mínimo viável (PMV); e d) elaboração de protótipos para validação de usabilidade. Resultados. : Foi evidenciada heterogeneidade significativa nos registros atuais e ausência de análise posterior dos dados. Foi desenvolvido um PVM que inclui um formulário de registro de quedas, visualização de medidas preventivas, perfis de usuário diferenciados e ferramentas educativas. O sistema foi validado internamente por cuidadores, gestores e profissionais de saúde em ILPI. Discussão. : O uso de ML permitiria a automação da análise de dados e a personalização de medidas preventivas. O design participativo e uma abordagem preventiva foram fundamentais para sua aceitabilidade. Conclusões. : O protótipo desenvolvido tem o potencial para otimizar o registro de quedas em ILPI, melhorar a prevenção e fortalecer o atendimento a idosos institucionalizados. [ABSTRACT FROM AUTHOR] |
| Copyright of MedUNAB is the property of Universidad Autonoma de Bucaramanga, UNAB and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | MedicLatina |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Introduction. : Falls in institutionalized older adults represent an underestimated public health problem associated with disability, dependence and mortality. In Chile, the absence of standardized records in long-term care facilities (LTCF) for older adults limits effective prevention. The objective of this study was to design a prototype of a digital fall registration system based on machine learning (ML) for its implementation in LTCF. Methodology. : The double diamond design methodology was used in four phases: a) identifying stakeholders and gathering information through qualitative interviews; b) analyzing causes and prioritizing ideas with an Analytical Hierarchy Process (AHP) and Pugh matrices; c) conceptual design and generating the minimum viable product (MVP), and d) developing prototypes for usability validation. Results. : There was evidence of great heterogeneity in the current records and a lack of subsequent data analysis. A MVP was developed, which includes a form for recording falls, a visualization of preventive measures, differentiated user profiles and educational tools. The system was internally validated by caregivers, managers and health care professionals in LTCF. Discussion. : Using ML would make it possible to automate data analysis and customize preventive measures. The participatory design and preventive approach were key to its acceptability. Conclusions. : The developed prototype has the potential to optimize how falls are recorded in LTCF, improve prevention and strengthen care for institutionalized older adults. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 01237047 |
| DOI: | 10.29375/01237047.5165 |
