VALIDATION AND RELIABILITY OF INERTIAL SENSORS TO MEASURE PELVIC ORIENTATION IN HEALTHY ADULTS

ABSTRACT Introduction In clinical practice, the degree of pelvic tilt is routinely evaluated because of its relationship to pelvic, spinal, and lower limb pathologies. The most common approach for assessing pelvic tilt in rehabilitation is visual observation and that is often qualitative in nature. Aside from radiographs, almost all quantitative techniques for measuring pelvic tilt in the sagittal plane require palpation of pelvic landmarks in static standing poses. Inertial sensors may provide a more versatile and reliable option for clinical pelvic tilt measurement. Purpose and Hypothesis The primary purpose of this study was to evaluate the intra-tester reliability of the measurement of pelvic tilt with an inertial measurement unit (IMU) system and compare that reliability to a palpation-based method. The secondary purpose of this study was to evaluate the concurrent validity of the IMU system compared to the palpation method when measuring pelvic tilt. The first hypothesis was that the intra-tester reliability for the IMU would be the same or greater than the palpation-based method. The second hypothesis of this study was that the IMU would have a good agreement to the palpation-based method. Methods A total of 29 healthy individuals (males n=14, females n=15) between the ages of 18-40 years were included in this study. Participants were asked to assume a comfortable standing position, sit on an inflated exercise ball, and assume a position with four-points of contact with both hands and both knees on the ground. All participants also performed sub-poses for standing, sitting, and four-point positions consisting of the preferred hip flexed for sitting and standing and extended for four-point positions. For the palpation method, the researcher palpated the anterior superior iliac spine (ASIS) and posterior superior iliac spine (PSIS) in each position and sub-pose. For the IMU method, an IMU sensor was attached with loop to an elastic belt wrapped around the lower back at the participant's sacral region. All palpation measurements were done on the side of the participant's preferred leg. Palpation data were collected using hand-held probes placed on the palpation points that were tracked using an 8-camera 3D motion capture system. Pelvic tilt was calculated as a primary variable for each position and sub-pose. As a secondary variable, pelvic width was also calculated from the palpation method. Pelvic width was used to assess repeatability of palpation method. An intra-class correlation (ICC₃‚₁) was used to assess the correlation of pelvic tilt measurements between positions during sitting, standing, and four-point positions for normal and knee raise poses estimated by IMU and palpation-based methods to assess test-retest reliability. Precision was calculated using the standard error of measurement (SEM) and SEM was then used to generate the minimal detectable difference (MDD) of pelvic tilt angle by IMU and palpation-based method. To investigate the secondary question, Bland-Altman plots were used to analyze the degree of agreement (i.e., concurrent validity) between pelvic angle measurement by IMU and palpation-based method. Results IMU measurements for pelvic tilt during all normal and knee raise sub-poses showed moderate to high intra-rater reliability (ICC= 0.71-0.85). Palpation measurements for pelvic tilt during all normal and knee raise sub-poses showed low to high intra-tester reliability (ICC= 0.27-0.84). For the IMU method, the SEM and MDD ranged from 5.2° to 8.0° and 8.4° to 12.8° respectively during all poses. For the palpation method, the SEM and MDD ranged from 3.5° to 7.4°and 7.1° to 11.8° respectively for all poses. Bland-Altman plots showed a bias between IMU and palpation methods for all poses that ranged from 8.8° to 14.8°. Conclusion An IMU, as implemented in this study (on a belt and wrap around the patient’s sacrum), is? a moderately reliable tool to help clinicians measure pelvic tilt in a clinical setting. The IMU is relatively easy to use and appears to be a more reliable tool as compared to the standard palpation approach