Publicación: Evaluación del deterioro funcional en la marcha del adulto mayor mediante un sistema de sensorización inercial y técnicas de aprendizaje automático
Fecha
2021-10-01
Autores
Editor/a
Director/a
Tutor/a
Coordinador/a
Prologuista
Revisor/a
Ilustrador/a
Derechos de acceso
Atribución-NoComercial-SinDerivadas 4.0 Internacional
info:eu-repo/semantics/openAccess
info:eu-repo/semantics/openAccess
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Nacional de Educación a Distancia (España). Escuela Técnica Superior de Ingeniería Informática. Departamento de Inteligencia Artificial
Resumen
Actualmente se está produciendo un aumento de la población envejecida en todo el mundo, pronosticando que en 2050 una de cada seis personas tendrá más de 65 años. Ahora bien, no todas las personas que llegan a edad avanzada cuentan con el grado de independencia que les permita llevar a cabo las actividades de la vida diaria de forma apropiada. La realización de dichas actividades permite determinar el riesgo de dependencia de las personas mayores, y por ende su nivel de autosuficiencia. El problema se presenta en aquellas personas, para las cuales, realizar las actividades de la vida diaria puede suponer un problema debido a su condición funcional. Las personas mayores que cuenten con una capacidad funcional reducida pueden sufrir accidentes, como las caídas no intencionadas, durante la realización de estas actividades poniendo en peligro su salud. Accidentes como las caídas pueden ocasionar lesiones, incapacidades, fracturas o incluso la muerte. Para evitar estos accidentes es conveniente evaluar la capacidad funcional del adulto mayor, con el fin de prevenir estos percances que puedan ocurrir durante la realización de las actividades de la vida diaria. Para realizar esta evaluación funcional han surgido, a lo largo de la historia, diversas pruebas clinimétricas que evalúan diferentes capacidades funcionales de la persona, como la escala de Berg o el test Tinetti. Estas pruebas obtienen una medida cuantitativa otorgada por el facultativo de la capacidad funcional del paciente, lo que permite anticiparse a los problemas que puedan ocurrir durante las actividades básicas e instrumentales. Uno de los principales problemas de estas pruebas es su carácter observacional y subjetivo, ya que, en la mayoría de los casos requieren que uno o varios profesionales sanitarios observen y evalúen al paciente bajo su punto de vista, lo que podría acarrear un sesgo de valoración de un profesional a otro. En este contexto, donde las técnicas de machine learning aportan un valor añadido, este Trabajo de Fin de Máster presenta el diseño y desarrollo de una aplicación que haga uso de diversos modelos de inteligencia artificial, que permitan evaluar automáticamente la capacidad funcional en la marcha de una persona mediante los datos obtenidos con sensores inerciales. Este trabajo realizará todas las fases de la ciencia de datos, desde la elaboración del conjunto de datos con sujetos reales utilizando sensores inerciales, pasando por el preprocesado, procesado, análisis, y analizar con el despliegue de los modelos en la aplicación diseñada e implementada.
There is currently an increase in the ageing population worldwide, with one in six people predicted to be over 65 by 2050. However, not all people who reach old age have the degree of independence that allows them to carry out activities of daily living appropriately. The performance of these activities determines the risk of dependency of older people, and thus their level of self-suficiency. The problem arises for those persons, for whom performing activities of daily living may be a problem due to their functional condition. Older people with reduced functional capacity may suffer accidents, such as unintentional falls, while performing these activities, endangering their health. Accidents such as falls can lead to injury, disability, fractures or even death. In order to avoid these accidents, it is advisable to assess the functional capacity of the older adult in order to prevent these mishaps that may occur during the performance of activities of daily living. In order to carry out this functional assessment, various clinimetric tests have emerged throughout history that evaluate different functional capacities of the person, such as the Berg scale or the Tinetti test. These tests provide a quantitative measure given by the physician of the patient’s functional capacity, which makes it possible to anticipate problems that may occur during basic and instrumental activities. One of the main problems with these tests is their observational and subjective nature, as in most cases they require one or more healthcare professionals to observe and assess the patient from their point of view, which could lead to a bias in assessment from one professional to another. In this context, where machine learning techniques provide added value, this Master’s Thesis presents the design and development of an application that makes use of various artificial intelligence models, which automatically evaluate the functional capacity in the gait of a person using data obtained with inertial sensors. This work will carry out all the phases of data science, from the elaboration of the data set with real subjects using inertial sensors, through pre-processing, processing, analysis, and ending with the deployment of the models in the designed and implemented application.
There is currently an increase in the ageing population worldwide, with one in six people predicted to be over 65 by 2050. However, not all people who reach old age have the degree of independence that allows them to carry out activities of daily living appropriately. The performance of these activities determines the risk of dependency of older people, and thus their level of self-suficiency. The problem arises for those persons, for whom performing activities of daily living may be a problem due to their functional condition. Older people with reduced functional capacity may suffer accidents, such as unintentional falls, while performing these activities, endangering their health. Accidents such as falls can lead to injury, disability, fractures or even death. In order to avoid these accidents, it is advisable to assess the functional capacity of the older adult in order to prevent these mishaps that may occur during the performance of activities of daily living. In order to carry out this functional assessment, various clinimetric tests have emerged throughout history that evaluate different functional capacities of the person, such as the Berg scale or the Tinetti test. These tests provide a quantitative measure given by the physician of the patient’s functional capacity, which makes it possible to anticipate problems that may occur during basic and instrumental activities. One of the main problems with these tests is their observational and subjective nature, as in most cases they require one or more healthcare professionals to observe and assess the patient from their point of view, which could lead to a bias in assessment from one professional to another. In this context, where machine learning techniques provide added value, this Master’s Thesis presents the design and development of an application that makes use of various artificial intelligence models, which automatically evaluate the functional capacity in the gait of a person using data obtained with inertial sensors. This work will carry out all the phases of data science, from the elaboration of the data set with real subjects using inertial sensors, through pre-processing, processing, analysis, and ending with the deployment of the models in the designed and implemented application.
Descripción
Categorías UNESCO
Palabras clave
Citación
Centro
Facultades y escuelas::E.T.S. de Ingeniería Informática
Departamento
Inteligencia Artificial