“…Single-leg sensor metrics were associated with surrogate measures of varus thrust, and midthigh adduction velocity was significantly associated with peak external knee adduction moment 3 IMU b (Trigno IM Sensors, Delsys Inc) on thigh, midshank, and distal shank Knee OA a (n=26) Costello et al, 2020 [33] Positive correlation between lateral thrust and change in medial meniscus extrusion 2 IMU c (WAA-010, ATR-Promotions) placed on tibia and foot Knee OA (n=44) Ishii et al, 2020 [34] Moderate correlation found between acceleration peak in IMU frame and KAM, values from shank IMU had strongest correlation 6 IMU c (TSND151, ATR-Promotions) on pelvis, sternum, shanks, and thighs Knee OA (n=22) Iwama et al, 2021 [35] High accuracy and repeatability of foot progression angle measures, and feedback effectiveness was similar between wearable and laboratory feedback setups 7 IMU b (MTw, Xsens Technologies BV) on pelvis, thighs, shanks, and feet Healthy (n=11) Karatsidis et al, 2018 [38] Two machine learning algorithms were highly accurate (R 2 approximately 0.95) in predicting KAM using IMU input 2 IMU c (DA14583, Dialog Semiconductor) on malleoli Healthy (n=12), knee OA (n=78) Wang et al, 2020 [36] Good correlation coefficients to discriminate between different foot progression angle walking conditions 2 IMU b (MTw Awinda, Xsens Technologies BV) on feet Healthy (n=5) Wouda et al, 2021 [37] Participants were able to respond to feedback during walking and adopt target foot progression angle conditions Single IMU b (custom-made) embedded in shoe sole Healthy (n=10) Xia et al, 2020 [39] a OA: osteoarthritis.…”