TFG | Support for the rehabilitation of stroke patients
Decision support system for the automatic definition and assignment of personalized rehabilitation routines for stroke patients. TFG developed by Sergio Martínez Cid and directed by David Vallejo Fernández and Cristian Gómez Portes. Project awarded in the II edition of the Turing Machine Awards 2022.
ABSTRACT:
Stroke or cerebrovascular accident is one of the leading causes of death and disability in the world. Stroke has a growing impact both socially and economically, and this impact increases in developing countries and in low-income sections of the population. Stroke affects both the cognitive abilities and the physical abilities of patients. In addition, patients need to carry out a rehabilitation process that usually lasts several months. Due to the impact of strokes, there are numerous computer systems that try to improve stroke prevention and rehabilitation processes. Specifically, there are projects dedicated to facilitating the execution of physical rehabilitation from home.
Rehabilitation from home allows to reduce costs related to the transport of the patient to the clinic and makes rehabilitation more accessible, but it generates challenges related to motivation and the correct execution of the exercises. The project detailed in this document is framed in the context of the development of a commercial system in the company Furious Koalas, development in which the author of the project has taken part. The system is a web application that aims to facilitate the rehabilitation of patients by guiding their movements and supporting the therapist in monitoring the patient's progress.
The project has consisted in the development of a decision support system, which is in charge of automatically defining physical rehabilitation routines for stroke patients. The project also includes integration into the system. The automatic definition of physical rehabilitation routines is an advantage for the therapist, since it reduces the time spent on defining routines. Thus, the quality time that the therapist spends with patients can be increased. Additionally, the decision support system includes the ability to generate explanations of suggested rehabilitation routines. This functionality falls within the field of Explainable Artificial Intelligence (XAI), which will be key to overcoming the ethical obstacles that prevent the integration of artificial intelligence in areas such as medicine.
