Search

Background: Three-dimensional gait analysis, supported by advanced sensor systems, is a crucial component in the rehabilitation assessment of post-stroke hemiplegic patients. However, the sensor data generated from such analyses are often complex and challenging to interpret in clinical practice, requiring significant time and complicated procedures. The Gait Deviation Index (GDI) serves as a simplified metric for quantifying the severity of pathological gait. Although isokinetic dynamometry, utilizing sophisticated sensors, is widely employed in muscle function assessment and rehabilitation, its application in gait analysis remains underexplored. Objective: This study aims to investigate the use of sensor-acquired isokinetic muscle strength data, combined with machine learning techniques, to predict the GDI in hemiplegic patients. This study utilizes data captured from sensors embedded in the Biodex dynamometry system and the Vicon 3D motion capture system, highlighting the integration of sensor technology in clinical gait analysis. Methods: This study was a cross-sectional, observational study that included a cohort of 150 post-stroke hemiplegic patients. The sensor data included measurements such as peak torque, peak torque/body weight, maximum work of repeated actions, coefficient of variation, average power, total work, acceleration time, deceleration time, range of motion, and average peak torque for both flexor and extensor muscles on the affected side at three angular velocities (60°/s, 90°/s, and 120°/s) using the Biodex System 4 Pro. The GDI was calculated using data from a Vicon 3D motion capture system. This study employed four machine learning models—Lasso Regression, Random Forest (RF), Support Vector regression (SVR), and BP Neural Network—to model and validate the sensor data. Model performance was evaluated using mean squared error (MSE), the coefficient of determination (R2), and mean absolute error (MAE). SHapley Additive exPlanations (SHAP) analysis was used to enhance model interpretability. Results: The RF model outperformed others in predicting GDI, with an MSE of 16.18, an R2 of 0.89, and an MAE of 2.99. In contrast, the Lasso Regression model yielded an MSE of 22.29, an R2 of 0.85, and an MAE of 3.71. The SVR model had an MSE of 31.58, an R2 of 0.82, and an MAE of 7.68, while the BP Neural Network model exhibited the poorest performance with an MSE of 50.38, an R2 of 0.79, and an MAE of 9.59. SHAP analysis identified the maximum work of repeated actions of the extensor muscles at 60°/s and 120°/s as the most critical sensor-derived features for predicting GDI, underscoring the importance of muscle strength metrics at varying speeds in rehabilitation assessments. Conclusions: This study highlights the potential of integrating advanced sensor technology with machine learning techniques in the analysis of complex clinical data. The developed GDI prediction model, based on sensor-acquired isokinetic dynamometry data, offers a novel, streamlined, and effective tool for assessing rehabilitation progress in post-stroke hemiplegic patients, with promising implications for broader clinical application.

During the polishing process of complex curved PMMA parts, the polishing force is an important factor affecting the surface quality and optical performance. In this paper, a force-controlled polishing device integrated into a machining center to maintain the polishing force is investigated. In order to achieve the real-time active control of the polishing force, the linear voice coil motor and force sensors are used for motion and measurement. A compact structure was designed to couple the linear motion of the voice coil motor with the rotation for polishing. The force-controlled polishing system with a high real-time hardware architecture was developed to perform complex curved polishing path movement with precise force control. Next, the polishing force between the device and the workpiece was analyzed to obtain the mathematical model of the device. Considering the impact during the approaching phase of polishing, a fuzzy PI controller was proposed to reduce the overshoot and response time. To implement the control method, the controller model was established on Simulink and the control system was developed based on TwinCAT 3 software with real-time computing capability. Finally, a polishing experiment involving a complex curved PMMA part was conducted by a force-controlled polishing device integrated into a five-axis machining center. The results show that the device can effectively maintain the polishing force to improve surface quality and optical performance.

Odonate is a new, intelligent three-dimensional gait analysis system based on binocular depth cameras and neural networks, but its accuracy has not been validated. Twenty-six healthy subjects and sixteen patients with post-stroke were recruited to investigate the validity and reliability of Odonate for gait analysis and examine its ability to discriminate abnormal gait patterns. The repeatability tests of different raters and different days showed great consistency. Compared with the results measured by Vicon, gait velocity, cadence, step length, cycle time, and sagittal hip and knee joint angles measured by Odonate showed high consistency, while the consistency of the gait phase division and the sagittal ankle joint angle was slightly lower. In addition, the stages with statistical differences between healthy subjects and patients during a gait cycle measured by the two systems were consistent. In conclusion, Odonate has excellent inter/intra-rater reliability, and has strong validity in measuring some spatiotemporal parameters and the sagittal joint angles, except the gait phase division and the ankle joint angle. Odonate is comparable to Vicon in its ability to identify abnormal gait patterns in patients with post-stroke. Therefore, Odonate has the potential to provide accessible and objective measurements for clinical gait assessment.