In-lab versus at-home activity recognition in ambulatory subjects with incomplete spinal cord injury
- PDF / 592,531 Bytes
- 6 Pages / 595.276 x 790.866 pts Page_size
- 8 Downloads / 202 Views
RESEARCH
Open Access
In-lab versus at-home activity recognition in ambulatory subjects with incomplete spinal cord injury Mark V. Albert1,2,3*, Yohannes Azeze1,4, Michael Courtois3 and Arun Jayaraman1,2
Abstract Background: Although commercially available activity trackers can aid in tracking therapy and recovery of patients, most devices perform poorly for patients with irregular movement patterns. Standard machine learning techniques can be applied on recorded accelerometer signals in order to classify the activities of ambulatory subjects with incomplete spinal cord injury in a way that is specific to this population and the location of the recording—at home or in the clinic. Methods: Subjects were instructed to perform a standardized set of movements while wearing a waist-worn accelerometer in the clinic and at-home. Activities included lying, sitting, standing, walking, wheeling, and stair climbing. Multiple classifiers and validation methods were used to quantify the ability of the machine learning techniques to distinguish the activities recorded in-lab or at-home. Results: In the lab, classifiers trained and tested using within-subject cross-validation provided an accuracy of 91. 6%. When the classifier was trained on data collected in the lab but tested on at home data, the accuracy fell to 54. 6% indicating distinct movement patterns between locations. However, the accuracy of the at-home classifications, when training the classifier with at-home data, improved to 85.9%. Conclusion: Individuals with unique movement patterns can benefit from using tailored activity recognition algorithms easily implemented using modern machine learning methods on collected movement data. Keywords: Activity recognition, Activity tracking, Incomplete spinal cord injury, At-home, Machine learning
Background Activity tracking can be performed using wearable sensors, which provide a wealth of information to encourage beneficial movement. Commercial activity tracking has gained immense popularity with a number of consumer devices available to help the general population track their fitness goals [1, 2]. Patients with motor disabilities, such as those with spinal cord injury, can benefit from activity tracking especially in therapeutic or clinical environments [3, 4]. Many of these devices, however, are not designed to work effectively for movement-impaired populations as they have not been validated in these populations. Precise * Correspondence: [email protected] 1 Max Nader Lab for Rehabilitation Technologies and Outcomes Research, Rehabilitation Institute of Chicago, Chicago, USA 2 Department of Computer Science, Loyola University Chicago, Chicago, USA Full list of author information is available at the end of the article
and automatic activity recognition has the potential to help create and evaluate individualized treatments plans, but more work must be done to improve the experience for movement-impaired patient populations. For individuals with spinal cord injuries, half of motor recovery occurs within the first few mon
Data Loading...
