The use of inertial sensors for the classification of rehabilitation exercises. Giggins, O., Sweeney, K. T., & Caulfield, B. In 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pages 2965–2968, August, 2014.
doi  abstract   bibtex   
The benefits of exercise in rehabilitation after orthopaedic surgery or following a musculoskeletal injury has been widely established. Within a hospital or clinical environment, adherence levels to rehabilitation exercise programs are high due to the supervision of the patient during the rehabilitation process. However, adherence levels drop significantly when patients are asked to perform the program at home. This paper describes the use of simple inertial sensors for the purpose of developing a biofeedback system to monitor adherence to rehabilitation programs. The results show that a single sensor can accurately distinguish between seven commonly prescribed rehabilitation exercises with accuracies between 93% and 95%. Results also show that the use of multiple sensor units does not significantly improve results therefore suggesting that a single sensor unit can be used as an input to an exercise biofeedback system.
@inproceedings{giggins_use_2014,
	title = {The use of inertial sensors for the classification of rehabilitation exercises},
	doi = {10.1109/EMBC.2014.6944245},
	abstract = {The benefits of exercise in rehabilitation after orthopaedic surgery or following a musculoskeletal injury has been widely established. Within a hospital or clinical environment, adherence levels to rehabilitation exercise programs are high due to the supervision of the patient during the rehabilitation process. However, adherence levels drop significantly when patients are asked to perform the program at home. This paper describes the use of simple inertial sensors for the purpose of developing a biofeedback system to monitor adherence to rehabilitation programs. The results show that a single sensor can accurately distinguish between seven commonly prescribed rehabilitation exercises with accuracies between 93\% and 95\%. Results also show that the use of multiple sensor units does not significantly improve results therefore suggesting that a single sensor unit can be used as an input to an exercise biofeedback system.},
	booktitle = {2014 36th {Annual} {International} {Conference} of the {IEEE} {Engineering} in {Medicine} and {Biology} {Society}},
	author = {Giggins, O. and Sweeney, K. T. and Caulfield, B.},
	month = aug,
	year = {2014},
	keywords = {Accuracy, Anzahl Sensoren, Biological control systems, Biosensors, Exercise, Exercise Therapy, Female, Hip, Humans, Joint Diseases, Knee, Male, Middle Aged, Sensitivity and Specificity, Sensor systems, adherence monitoring, biomedical equipment, exercise biofeedback system, inertial sensor, musculoskeletal injury, orthopaedic surgery, patient monitoring, patient rehabilitation, patient treatment, rehabilitation exercise classification},
	pages = {2965--2968}
}

Downloads: 0