Detection of the Intention to Grasp during Reaching in Stroke Using Inertial Sensing
| Autor: | Peter H. Veltink, A. L. van Ommeren, B. Sawaryn, Jaap H. Buurke, Johan S. Rietman, Gerdienke B. Prange-Lasonder |
|---|---|
| Přispěvatelé: | Biomedical Signals and Systems, Biomechanical Engineering |
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: |
Male
030506 rehabilitation Support Vector Machine Activities of daily living Inertial frame of reference Computer science Biomedical Engineering Wearable computer Intention Thumb 03 medical and health sciences 0302 clinical medicine Activities of Daily Living assistive technology Internal Medicine medicine Humans Computer vision Survivors Stroke Aged Hand Strength business.industry General Neuroscience Rehabilitation GRASP Stroke Rehabilitation Robotics Middle Aged Self-Help Devices medicine.disease Biomechanical Phenomena Support vector machine medicine.anatomical_structure machine learning Feature (computer vision) soft-robotic glove grasp intention Female Artificial intelligence 0305 other medical science business Psychomotor Performance 030217 neurology & neurosurgery inertial sensing |
| Zdroj: | IEEE transactions on neural systems and rehabilitation engineering, 27(10):8844826, 2128-2134. IEEE |
| ISSN: | 1534-4320 |
| Popis: | To support stroke survivors in activities of daily living, wearable soft-robotic gloves are being developed. An essential feature for use in daily life is detection of movement intent to trigger actuation without substantial delays. To increase efficacy, the intention to grasp should be detected as soon as possible, while other movements are not detected instead. Therefore, the possibilities to classify reach and grasp movements of stroke survivors, and to detect the intention of grasp movements, were investigated using inertial sensing. Hand and wrist movements of 10 stroke survivors were analyzed during reach and grasp movements using inertial sensing and a Support Vector Machine classifier. The highest mean accuracies of 96.8% and 83.3% were achieved for single- and multi-user classification respectively. Accuracies up to 90% were achieved when using 80% of the movement length, or even only 50% of the movement length after choosing the optimal kernel per person. This would allow for an earlier detection of 300-750ms, but at the expense of accuracy. In conclusion, inertial sensing combined with the Support Vector Machine classifier is a promising method for actuation of grasp-supporting devices to aid stroke survivors in activities of daily living. Online implementation should be investigated in future research. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|