Alfikri, Zakiy F
(2023)
A Comprehensive mHealth System for Chronic Low Back Pain Assessment: Development, Evaluation, and Exploration for Future Works.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
Phenotyping chronic low back pain (CLBP) is essential for developing personalized and adaptive treatments for CLBP. To achieve this, a large amount of CLBP assessment data is needed. In this study, an mHealth system was developed for CLBP assessment. The system consists of an in-clinic app, at-home app, clinician portal, backend, and database, and focuses on collecting biomechanical and behavioral assessment data as part of extensive multifactorial data needed for CLBP phenotyping. The mHealth system was able to collect CLBP assessment data effectively and efficiently from both the structured in-clinic assessment and the assessment in the patients’ daily life settings.
Usability evaluations were conducted to assess the usability of the in-clinic and at-home apps. Two usability evaluations were conducted for the in-clinic app, and several updates and revisions were made to address identified usability issues. In the last evaluation, the in-clinic app received a high usability score of 6.00 (SD=1.15). Meanwhile, for the at-home app, five usability evaluations, involving 337 CLBP patients, were conducted. Several updates and revisions were made to address the usability issues identified. In the last usability evaluation, the at-home app received a high usability score of 6.24 (SD=1.37).
Furthermore, two exploratory works for future direction were conducted in this study. The first was an investigation of the correlation between the perceived activity level that patients reported in EMA and the activity level calculated from accelerometer data from the kinematics sensors. The overall correlation was found to be weak, ranging from 0.095 to 0.260 (mean=0.194, SD=0.054). Even though the overall correlation was weak, the correlation of activity level from sensor data and perceived activity level from 37.5% of CLBP patients was found to be strong. Using the five most accurate activity level representations, the average score for the correlation was 0.716 (SD=0.081), suggesting that some CLBP patients may have a better perception of their activity level. The other exploratory work done in this study was the development of a dataset builder component that was successfully be used to label motion data based on the videos recorded during the in-clinic assessment.
Share
| Citation/Export: |
|
| Social Networking: |
|
Details
| Item Type: |
University of Pittsburgh ETD
|
| Status: |
Unpublished |
| Creators/Authors: |
|
| ETD Committee: |
|
| Date: |
6 June 2023 |
| Date Type: |
Publication |
| Defense Date: |
13 March 2023 |
| Approval Date: |
6 June 2023 |
| Submission Date: |
22 March 2023 |
| Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
| Number of Pages: |
213 |
| Institution: |
University of Pittsburgh |
| Schools and Programs: |
School of Health and Rehabilitation Sciences > Health Information Management |
| Degree: |
PhD - Doctor of Philosophy |
| Thesis Type: |
Doctoral Dissertation |
| Refereed: |
Yes |
| Uncontrolled Keywords: |
mHealth, app, chronic low back pain, assessment, development, usability, physical activity level, EMA, kinematics sensor, machine learning, dataset builder |
| Date Deposited: |
06 Jun 2023 13:51 |
| Last Modified: |
06 Jun 2023 13:51 |
| URI: |
http://d-scholarship.pitt.edu/id/eprint/44311 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](/images/feed-icon-32x32.png){kind=icon}