The purpose of this investigation was to determine the accuracy of Single Leg Jump focused metrics measured by Plantiga insoles.
The experiments were conducted with participants wearing sensor insoles, while performing different types of Single Leg Jumps on a dual force plate platform (Kistler). The data was collected and processed for both the insoles and force plates and then single leg jump height and distance metrics were computed and compared.
For the jump height metric, linear regression showed a coefficient of determination (r-squared) value of 0.99 and a slope of 1. For the jump distance metric, regression showed a coefficient of determination of 0.99 and a slope of 1.04. These results show that Plantiga sensor insoles provide a reliable and accurate way of measuring jump distances and height, while being lightweight and portable.
Among different Single Leg Jump metrics measured by Plantiga’s insole system, two were validated in this report. They are highlighted below:
Jumping to gain height creates larger magnitudes of work at the knee joints and targets primarily knee extensors. Jump height tests can be used to determine functional deficits of the quadriceps complex between the left and right leg.
Jumping for distance requires complex hip and trunk movements to maintain postural stability and creates larger work at the hip and ankle joint when compared to jumping for height. Jump distance tests can be used to determine functional deficits in either the hip or ankle complex.
Plantiga insoles utilize a state-of-the-art custom built Machine Learned Model to detect and classify movement data into segments with millisecond accuracy. The model is able to detect when the foot becomes airborne (takeoff) and when it touches the ground (landing). This takeoff-landing information is then fed into the Jump Detection Model to compute jump metrics. For more information on the ML Model, see Ground Interaction Algorithm Validation.
Single Leg Jump data were collected from 14 individuals. A total of 195 height based jumps were analyzed while a total of 120 distance based jumps were analyzed. Each participant was asked to wear the insoles and jump on a dual force plate platform (Kistler) with right and left leg in an alternating manner. The force plate sampled the signal at a higher rate of 1000 Hz compared to the insole sampling rate of 500 Hz. Once the two signals were synchronized, the force plate data was downsampled to match the insole sampling frequency of 500 Hz. The force plate data was then used to compute the reference jump height.
To get a reference jump distance to compare against insole computed distance, participants were asked to jump forward from a fixed starting point according to their comfort level, while the jump was recorded via a high frame-per-second camera. After a jump was completed, video was used to pinpoint the landing point and then the total distance was measured by a measuring tape.
Linear regression analysis showed a coefficient of determination (r-squared) value of 0.99 with a slope of 1. Further, signed error statistics were observed with a median value of 0 and an IQR (InterQuartile Range) of 4.8%. These results are shown in Figure 1 and 2 respectively.
Figure 1. Linear Regression Analysis for Single Leg Jump Height.
Figure 2. Error Statistics for Single Leg Jump Height.
Singed Error (cm): Median: 0.0, Q1: -0.09, Q3: 0.10, IQR: 0.19.
Percent Error (%) : Median: 0.0, Q1: -2.37, Q3: 2.43, IQR: 4.8 %