Wearable Sensors Generate New Measured Outcomes for Cervical Myelopathy

Introduction: Clinical evaluation of CSM is limited as Hoffman’s sign, Romberg testing and Tandem Gait are largely binary, making deterioration or improvement difficult to document accurately. We report on the role of a single wearable sensor in the evaluation of CSM. This study investigates the use of AI to identify new metrics that enhance our ability to measure disease progression and response to treatment.

Methods: Patients scheduled for surgical treatment of CSM underwent in-office and 24-hour continuous at-home data collection using a single wearable sensor. In-office testing consisted of Standing, Tandem Gait, Timed Up & Go (TUG). Testing was repeated 6-months post-op.

Results: Tandem gait revealed significant improvement in mean AP sway (p < 0.007) and max AP sway (p < 0.015) after surgery reflecting quantification of the standard clinical exam, but additionally a stronger correlation with speed of gait (p < 0.000) and a decrease in the initial impact as the foot hits the ground at each step (Initial Peak Acceleration, p=0.001). At-home testing demonstrated decreased turns during sleep (p < 0.019) and increased hours of sleep (p < 0.032).

Conclusion : Analysis of wearable sensor data using AI identifies new metrics with the potential to more accurately assess pre-op and post-op function in patients with CSM. Previously unreported pre-op to 6-month post-op changes were seen in speed of gait and ground impact force during Tandem Gait. These metrics were more sensitive as compared to the normal ant/post and lateral sway assessment. 24-hour sensor data showed marked improvement in sleep quality with decreased number of turns and increased overall hours of sleep post-op.
This study suggests that wearable sensor data will be a viable source for quantifiable data with the potential to guide treatment for patients with CSM. This capability is based partly on better quantification of existing binary measures, but also on the unique quality of AI analysis to identify unanticipated patterns within the data.