Development of SARA-Q: An Automated Assessment System Based on SARA for Enhanced Precision in Evaluating Motor Coordination

(Accepted to International Congress for Ataxia Research 2024)

​

By Eric Lee et. Al.

​

Abstract:

​

The Scale for the Assessment and Rating of Ataxia (SARA) is frequently employed in clinical evaluations. However, challenges arise when assessments must be carried out outside of a clinical setting or without an examiner present. These limitations present challenges in tracking daily variations and conducting precise analyses of outcomes. Additionally, the subjective nature of the examiner's assessment could impact objectivity and quantitative evaluation. In response to these challenges, we designed and implemented SARA-Q, an automated assessment platform for ataxia based on the SARA assessment tool.

​

SARA-Q employs a method where patients draw on a computer touchscreen, capturing x and y coordinates for analysis, thereby enabling a more precise evaluation of motor coordination. This app focuses on two upper limb assessment items: finger tracking and the finger-to-nose test. SARA-Q enhances the accuracy and reliability of assessments through various quantitative metrics, including deviation area analysis, reverse movement analysis, discontinuity analysis, test area overshoot analysis, tremor frequency analysis, maximum and average rotational amplitude analysis, and modulation speed analysis. Furthermore, SARA-Q allows assessments to be conducted by patients alone or with minimal caregiver assistance, facilitating data collection in various settings within a short time frame. In a beta test conducted on 10 brain injury patients at Chungbuk National University Hospital, it was confirmed that SARA-Q could more precisely differentiate between patients who were classified with the same score on the traditional SARA test. This was accomplished by employing precise data measured in increments of one-thousandth of a second and one-thousandth of a centimeter.

 

Additionally, the integration of AI technology is in progress to provide a personalized testing environment and to expand the system from simple assessments to rehabilitation exercises.