A Shoe-Integrated Sensor System for Long- Term Center of Pressure Evaluation

Abstract

In clinical, the center of pressure (CoP) is commonly used for accessing the stability of a person's postural control, which is highly associated with various neurological diseases and movement disorders such as Alzheimer's disease, Parkinson's disease, chronic ankle instability. Such a disease usually has a long development or rehabilitation process which requires long-term CoP monitoring. The current CoP evaluation process does not meet the requirement, as it is often complicated and expensive through either the lab-based equipment or the clinical evaluation procedure. Different wearable sensor-based systems with less cost and restrictions have emerged, but their way of CoP calculation requires deliberate calibration of the positions of their sensors, which are not feasible in daily CoP monitoring. In this study, we developed a longterm CoP monitoring system in a smart-shoe form. First, a thin and flexible smart insole with optimal sensor locations was designed to be compact and energy sufficient for a whole-day usage. Then, a user-friendly app on the smartphone with a cloud-based data managing system was developed for applications in both clinical and home environments. Additionally, a simplified CoP estimation model was created without the need for calibration. Lastly, a machine learning-based human activity recognition method was incorporated to make the CoP detection process more automatic. Through a thorough validation test with the clinical level lab equipment, our system can generate the CoP measurements with high accuracy.

The Application of Wearable Sensors and Machine Learning Algorithms in Rehabilitation Training: A Systematic Review

“…For different types of diseases, because of the different wearable sensors used and the different collected data types, the selection of features was also very different. In an article written on studying the walking gait of patients, the mean and standard deviation values of all the pressure data received by plantar pressure sensors were used as features [ 51 ]. There are also gait lines, regional pressures, gait phases, accelerations, step lengths, and joint angles that combine leg and plantar wearable sensor data as features [ 69 ].…”

“…Combining MMG signals can effectively improve the accuracy of symptom classification. For the gait research of Parkinson’s disease patients, Guo et al [ 51 ] used plantar pressure sensors to efficiently collect patient’s plantar pressure data, and the selection of low-power sensors can effectively extend the daily monitoring time. From the perspective of users, Han et al [ 57 ] selected a lighter IMU, which could reduce the patient’s exercise burden and ensure the completion of rehabilitation training.…”

“…An et al [ 63 ] trained the model by adding data recorded under different conditions, which can also effectively improve the accuracy of the model. Guo et al [ 51 ] applied the collected data features to a variety of machine learning algorithms, and the accuracy rates were generally similar; however, the accuracy of the neural network model was the highest, indicating that the amount of information collected by sensors can affect the accuracy of the model.…”

“…The method provides ideas for helping researchers and healthcare professionals analyze the complex movements of patients during their rehabilitation. Guo et al [ 51 ] aimed at the rehabilitation of Alzheimer’s, Parkinson’s, and other diseases, because the main rehabilitation training for such diseases lies in daily walking activity; therefore, a smart insole was used to monitor patients’ everyday walking activity. The design of such insoles has good development prospects for the rehabilitation of patients at home and in the community.…”

“…In terms of the selection of experimental subjects, the experimental participants presented in some papers [ 48 , 50 , 53 , 54 , 56 , 57 , 58 , 60 , 65 , 69 , 70 , 71 , 73 , 74 , 75 ] selected disease patients or a combination of disease patients and healthy participants for the experimental research. Another approach [ 42 , 46 , 47 , 49 , 51 , 55 , 59 , 62 , 66 , 67 , 68 , 72 , 76 ] was to recruit disease-simulated subjects to imitate patients for exercise experiments. There was a certain gap between the information collected by simulated subjects and the real data of patients, and it was difficult to guarantee the authenticity and validity of the research results.…”