ABSTRACT
Wearable sensors facilitate running kinematics analysis of joint kinematics in real running environments. The use of a few sensors or, ideally, a single inertial measurement unit (IMU) is preferable for accurate gait analysis. This study aimed to use a convolutional neural network (CNN) to predict level-ground running kinematics (measured by four IMUs on the lower extremities) by using treadmill running kinematics training data measured using a single IMU on the anteromedial side of the right tibia and to compare the performance of level-ground running kinematics predictions between raw accelerometer and gyroscope data. The CNN model performed regression for intraparticipant and interparticipant scenarios and predicted running kinematics. Ten recreational runners were recruited. Accelerometer and gyroscope data were collected. Intraparticipant and interparticipant R2 values of actual and predicted running kinematics ranged from 0.85 to 0.96 and from 0.7 to 0.92, respectively. Normalized root mean squared error values of actual and predicted running kinematics ranged from 3.6% to 10.8% and from 7.4% to 10.8% in intraparticipant and interparticipant tests, respectively. Kinematics predictions in the sagittal plane were found to be better for the knee joint than for the hip joint, and predictions using the gyroscope as the regressor were demonstrated to be significantly better than those using the accelerometer as the regressor.
Subject(s)
Running , Wearable Electronic Devices , Accelerometry , Biomechanical Phenomena , Exercise TestABSTRACT
Significant correlations for bone mineral density and bone microstructure between spinal and non-spinal skeletal sites (distal radius and proximal femur) in adolescent idiopathic scoliosis (AIS) patients were observed, indicating that proximal femoral DXA and distal radial HR-pQCT could provide valid clinical assessments in patients with AIS. PURPOSE: Low bone mass is an important feature of adolescent idiopathic scoliosis (AIS), which is a complex 3D spinal deformity that affects girls during puberty. However, no clinical imaging modality is suitable for regular monitoring on their spinal bone qualities in rapid growth period. Therefore, we investigated whether bone mineral density (BMD) and bone microstructure at non-spinal sites correlated with BMD and mechanical property in the spine in AIS patients. METHODS: Thirty-two AIS girls (16.7 ± 3.5 years old with mean Cobb angle of 67 ± 11°) who underwent pre-operative spine CT examination for navigation surgery were recruited. Volumetric BMD (vBMD) of lumbar spine (LS) was measured by quantitative computed tomography (QCT), vBMD and bone microstructure of distal radius (DR) by high-resolution peripheral QCT (HR-pQCT) and areal BMDs of total hip (TH) and femoral necks (FN) by dual-energy X-ray absorptiometry (DXA). Biomechanical properties of the DR and LS were estimated by finite element analysis (FEA). Pearson correlation was performed to study the correlation between bone parameters at these three sites. RESULTS: LS vBMD correlated significantly with both FN and TH aBMD (R = 0.663-0.725, both p < 0.01) and with DR microstructural parameters (R = 0.380-0.576, all p < 0.05). Mechanical properties of LS and DR were also correlated (R = 0.398, p = 0.039). CONCLUSIONS: Bone measurement at proximal femur and distal radius could provide an additional predictive power in estimating the bone changes at spine, which is the primary site of deformity in AIS patients. Our result indicated that DXA and HR-pQCT could provide a valid surrogate for spine bone measurements in AIS patients.