Prediction of Lower Extremity Multi-Joint Angles during Overground Walking by Using a Single IMU with a Low Frequency Based on an LSTM Recurrent Neural Network

Abstract

The joint angle during gait is an important indicator, such as injury risk index, rehabilitation status evaluation, etc. To analyze gait, inertial measurement unit (IMU) sensors have been used in studies and continuously developed; however, they are difficult to utilize in daily life because of the inconvenience of having to attach multiple sensors together and the difficulty of long-term use due to the battery consumption required for high data sampling rates. To overcome these problems, this study propose a multi-joint angle estimation method based on a long short-term memory (LSTM) recurrent neural network with a single low-frequency (23 Hz) IMU sensor. IMU sensor data attached to the lateral shank were measured during overground walking at a self-selected speed for 30 healthy young persons. The results show a comparatively good accuracy level, similar to previous studies using high-frequency IMU sensors. Compared to the reference results obtained from the motion capture system, the estimated angle coefficient of determination (R2) is greater than 0.74, and the root mean square error and normalized root mean square error (NRMSE) are less than 7° and 9.87%, respectively. The knee joint showed the best estimation performance in terms of the NRMSE and R2 among the hip, knee, and ankle joints.

Effect of Foot Orthoses on Angular Velocity of Feet

“…In sports and biomechanics, a recommended IMU sample rate is at least 20 Hz [19,20] with a 20 Hz cut-off frequency for lowpass filters observed to enhance prediction accuracy [10]. Florenciano et al [9] have reported that a sampling rate of 30 Hz is sufficient to capture the foot movement when running at low speed, and the reported results are consistent with the other literature [3,[21][22][23].…”

“…One of the limitations of this study relates to the sampling frequency used to collect the data. The sampling frequency used in this study has already been used in the past [3,9,13,[21][22][23], but it inevitably limits the running speed of the subjects. Whether these results are valid for higher speeds, where the impact of the feet and the movement itself will be different, needs to be further investigated.…”

Enhancing accuracy and convenience of golf swing tracking with a wrist-worn single inertial sensor

“…The collected data were filtered at a cut-off frequency of 20 Hz using a Butterworth 10th order low-pass filter. The order and cut-off frequency of the filter were selected to preserve the significant power of the acceleration and angular velocity signal for the fast motion of the golf swing while minimizing the power and peak loss due to filtering 9 , 54 – 57 . Cubic spline interpolation was performed using ten sample data points before and after the clipping segments from the original IMU data for correcting the clipping.…”

Criterion Validity of Linear Accelerations Measured with Low-Sampling-Frequency Accelerometers during Overground Walking in Elderly Patients with Knee Osteoarthritis

“…Moreover, a higher sampling frequency results in higher energy consumption and shorter battery lifetime and requires more expensive, complex, and bulky sensors. Accordingly, several AI algorithms have recently been proposed to increase the accuracy of the low sampling frequency of IMUs [ 32 , 33 ]. It has been demonstrated by previous studies that 98% of the walking signal’s power is below 10 Hz [ 34 ] and significantly below 5 Hz [ 35 ].…”