Citas bibligráficas
Mendez, P., (2024). Aplicación móvil de alertas de accidentes caseros en adultos mayores basado en modelos de Deep Learning [Universidad de Lima]. https://hdl.handle.net/20.500.12724/20764
Mendez, P., Aplicación móvil de alertas de accidentes caseros en adultos mayores basado en modelos de Deep Learning []. PE: Universidad de Lima; 2024. https://hdl.handle.net/20.500.12724/20764
@misc{renati/234554,
title = "Aplicación móvil de alertas de accidentes caseros en adultos mayores basado en modelos de Deep Learning",
author = "Mendez Avila, Pelmef Roemer",
publisher = "Universidad de Lima",
year = "2024"
}
In recent years, technology related to health has undergone a remarkable evolution in various fields of study, allowing people to become more integrated into it in their daily lives. In addition, the demographic growth of older adults is perceived and has the tendency to continue to grow, of the older adult population in Peru, a certain percentage, about 38.4% according to previous studies, lives alone, so the probability of any risk event, accident or other is much more likely and would affect the lifestyle, health and well-being. On the other hand, many of the older adults suffer from vulnerability, i.e. they have some kind of disease that needs frequent monitoring and care, such as heart disease, respiratory diseases, among others, or they suffered from some kind of accident that needs assistance. Given the current situation of older adults living alone, a challenge arises in monitoring the well-being and health of those who are alone and/or vulnerable to illness or accidents. This work proposes a solution to the problem, where family members or people who are watching over the elderly can monitor their well-being without having to affect their privacy or be watching them at all times. It seeks the correct classification and interpretation of falls, fainting and similar accidents that the person may suffer, through a convolutional network (neural network responsible for the detection of actions and objects in images and / or videos) YOLOv7, in this case the detection of position of people is used to obtain the corresponding coordinates for the calculation of height and width of the subject recognized and can calculate the time at which the subject is in a position recognized as a fall, then the notification is made to the interested persons who are in charge of the adult. This system aims for little human intervention in the monitoring of well-being and a short response time at the moment of detecting and notifying the identified situation.
IMPORTANTE
La información contenida en este registro es de entera responsabilidad de la universidad, institución o escuela de educación superior que administra el repositorio académico digital donde se encuentra el trabajo de investigación y/o proyecto, los cuales son conducentes a optar títulos profesionales y grados académicos. SUNEDU no se hace responsable por los contenidos accesibles a través del Registro Nacional de Trabajos de Investigación – RENATI.