Robustness of CGM2.4 medial-lateral intermediate cuneiform marker misplacement on forefoot-to-rearfoot angles

A two-segment foot model has been implemented in the Conventional Gait Model 2 (CGM2.4) 1,2 allowing clinicians to explore the forefoot-to-rearfoot kinematics with a minimum of markers, compared to the Oxford foot model3. Multi-segment foot models are particularly prone to marker misplacement, because segments and marker-to-marker distances are small, with between tester variability of around 5 mm4,5. Mainly the intermediate cuneiform (TOE) marker might be prone sensitive to marker misplacement as it is linked to both the forefoot and hindfoot segments in the CGM2.4 model. This study aimed to evaluate the robustness of the CGM2.4 intermediate cuneiform marker placement on forefoot-to-rearfoot angles. Fourteen healthy adults volunteered to participate in this study (7 males and 7 females; age: 31.1±9.8 years, height: 172.6±9.2 cm, mass: 67.7±9.9 kg, foot length: 25.4±1.8 kg). Participants completed five barefoot walking trials at a self-selected speed, where forefoot-to-rearfoot kinematics were recorded at 100 Hz with the CGM2.4 marker set (marker diameter: 14 mm) and an 8-camera T40 Vicon motion capture system. One trained tester performed the marker placement according to the guidelines 2. Post data collection, four virtual markers were systematically created in Vicon ProCalc in both the static and dynamic trials. More specific, the TOE markers original position was moved 2.5 and 5 mm in the medial and lateral direction, respectively, in the foot segment’s transverse local coordinate system. Forefoot-to-rearfoot kinematics were recalculated from each of the four virtual marker positions. Root Mean Square Error (RMSE) between the forefoot-to-rearfoot angles obtained from the original TOE and four virtual markers, respectively, was calculated to evaluate the robustness of TOE marker placement. Only data from the right foot was analysed. The RMSE was largest in the transverse plane, where marker misplacement of 2.5 and 5 mm resulted in average RMSEs between 3.0±0.5 and 6.1±0.9° adduction/abduction (Fig. 1). Whilst RMSE values between 1.3±0.3 and 2.7±0.5° were observed in the frontal plane. RMSE in the sagittal plane was unchanged. Range of motion data was consistent across TOE marker positions.Download : Download high-res image (193KB)Download : Download full-size image Fig. 1. Average gait cycle forefoot-to-rearfoot kinematics and RMSE (mean±SD) for the different TOE marker positions. Understanding the consequence of CGM2.4 marker misplacement is important to improve the reliability and application in clinical decision-making. The RMSE observed for 5 mm medial/lateral misplacement (6.1±0.9°) in this study,was above the 5° limit of clinical acceptability previously reported for lower-limb kinematics 6. In conclusion, forefoot-to-rearfoot gait kinematics of the CGM2.4 is prone to medial/lateral cuneiform (TOE) marker placement error, particularly in the transverse plane. Thus, clinicians should examine the absolute forefoot-to-rearfoot angles with caution. However, values of dynamic range of motion during gait seems to be more robust to marker misplacement.