Publicação
A biofeedback strategy for human postural sway control in a smart walker
| Resumo: | Older adults and patients with common neurological diseases are characterised to have a very high movement variability and sway, with an annual incidence of facing falling events rising to 60-80%. The cerebellum is the brain structure responsible for motor functioning, balance, and gait, which damage leads to ataxia. To take over these disturbances, rehabilitation has the goal to achieve an optimal outcome for each individual patient mediated by neuroplasticity processes. Biofeedback systems have been proving to increase the efficiency of the conventional rehabilitation therapies by providing biological information to patients in real-time, allowing their self-correction. Wear able and low-cost sensors, such as inertial measurement unit (IMU), have demonstrated special interest as they can measure the sway in medio-lateral (ML) and antero-posterior (AP) directions in different body locations, are easy to use, accessible to purchase, can monitor balance anywhere with no need of a specific environment, and may improve the patients’ motivation along the treatment. It is notable the lack of biofeedback systems incorporated in smart walkers (SW). This dissertation aims to design and develop a real-time visual and audio biofeedback system in the WALKit SW as a complementary rehabilitation tool. For such, several sensor fusion algorithms were tested, validated and analysed, being the Complementary filter the most suitable, both in terms of root mean squared error (RMSE) and computing time. An experiment with healthy participants, with the goal of defining threshold values to be applied in the biofeedback system, was performed. Data followed the literature in the sense that these values should be specific for each person, so the proposed strategy takes an individual-selection method to satisfy this condition. With this system, it is expected to improve the WALKit SW functioning, giving more motivation and autonomy to the user on self-correcting his movements and posture, possibly leading to a more efficient recovery. However, this strategy needs to be validated among patients with balance disorders and its long-term effects. |
|---|---|
| Autores principais: | Costa, Mariana Brito Palma Campos |
| Assunto: | Balance Biofeedback CoM Euler angles Gait IMU Posture Smart walker Andarilho inteligente Ângulos de Euler Balanço Biofeedback CdM Marcha Postura Sensores inerciais |
| Ano: | 2022 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade do Minho |
| Idioma: | inglês |
| Origem: | RepositóriUM - Universidade do Minho |
| Resumo: | Older adults and patients with common neurological diseases are characterised to have a very high movement variability and sway, with an annual incidence of facing falling events rising to 60-80%. The cerebellum is the brain structure responsible for motor functioning, balance, and gait, which damage leads to ataxia. To take over these disturbances, rehabilitation has the goal to achieve an optimal outcome for each individual patient mediated by neuroplasticity processes. Biofeedback systems have been proving to increase the efficiency of the conventional rehabilitation therapies by providing biological information to patients in real-time, allowing their self-correction. Wear able and low-cost sensors, such as inertial measurement unit (IMU), have demonstrated special interest as they can measure the sway in medio-lateral (ML) and antero-posterior (AP) directions in different body locations, are easy to use, accessible to purchase, can monitor balance anywhere with no need of a specific environment, and may improve the patients’ motivation along the treatment. It is notable the lack of biofeedback systems incorporated in smart walkers (SW). This dissertation aims to design and develop a real-time visual and audio biofeedback system in the WALKit SW as a complementary rehabilitation tool. For such, several sensor fusion algorithms were tested, validated and analysed, being the Complementary filter the most suitable, both in terms of root mean squared error (RMSE) and computing time. An experiment with healthy participants, with the goal of defining threshold values to be applied in the biofeedback system, was performed. Data followed the literature in the sense that these values should be specific for each person, so the proposed strategy takes an individual-selection method to satisfy this condition. With this system, it is expected to improve the WALKit SW functioning, giving more motivation and autonomy to the user on self-correcting his movements and posture, possibly leading to a more efficient recovery. However, this strategy needs to be validated among patients with balance disorders and its long-term effects. |
|---|