Posture Recognition Method for Caregivers during Postural Change of a Patient on a Bed using Wearable Sensors

Abstract

Caregivers experience lower back pain due to their awkward postures while handling patients. Therefore, a monitoring system to supervise caregivers' postures using wearable sensors is being developed. This study proposed a postural recognition method for caregivers during postural change while handling a patient on a bed. The proposed method recognizes foot positions and arm movements by a machine learning algorithm using inertial data on the trunk and foot pressure data obtained from wearable sensors. An experiment was conducted to evaluate whether the proposed method could recognize three foot positions and three arm movements. Participants provided postural change for a simulated patient on a bed (patient: supine to lateral recumbent) under nine conditions, including different combinations of the three foot positions and three arm movements; the experiment was repeated ten times for each condition. Experimental results showed that the proposed method using an artificial neural network with all features obtained from an inertial measurement unit and insole pressure sensors could recognize arm movements and foot positions with an accuracy of approximately 0.75 and 0.97, respectively. These results suggest that the proposed method can be used in a monitoring system tracking a caregiver's posture.

Foot Position Measurement during Assistive Motion for Sit-to-Stand Using a Single Inertial Sensor and Shoe-Type Force Sensors

“…In addition, inertial and force data were affected by the changing foot position ( Figure 4 and Figure 5 ). The inertial sensor on the trunk could perceive differences in the foot position given that trunk movement is affected by foot position during manual handling [ 19 , 51 ]. Moreover, the shoe-type force sensors could measure the force changes caused by the foot position given that the force distribution on the insole changes based on the foot position during manual handling [ 25 ].…”

“…The extracted parameters included the mean, standard deviation, skewness, kurtosis, maximum, minimum, and root mean square of the data in each measurement. These parameters were determined based on previous studies using machine learning algorithm and wearable sensors [ 19 , 24 , 25 ]. These seven parameters were calculated for 3-axes accelerations, 3-axes gyro, and four force values (front and rear of each foot); thus, a total of 70 features were calculated for each trial.…”

“…To address the aforementioned concerns, we have been developing a method for measuring foot position using a single inertial sensor on trunk and shoe-type force sensors [ 19 ]. Previous studies have shown that the use of an inertial sensor on the trunk and shoe-type force sensors can be considered an effective method for measuring foot position given the relationship between trunk movement and foot position during manual handling [ 9 , 20 , 21 , 22 , 23 ].…”

“…Lee et al revealed that the COP shifted to the left side when the right foot was placed forward [ 23 ]. Moreover, previous studies have shown that shoe-type force sensors could be used to recognize occupational postures, including different foot positions [ 24 , 25 ], and to monitor trunk movement, lumbar load, and arm movement during manual handling [ 19 , 25 , 26 , 27 ]. From these findings, we hypothesized that foot position can be estimated on the basis of the distribution of the ground reaction force on the foot that is obtained from shoe-type force sensors.…”

“…A single inertial sensor on the trunk and shoe-type force sensors could not measure the quantitative foot position [ 19 ]. Thus, the current study utilized machine learning-based regression algorithms for the indirect wearable measurement of quantitative foot position.…”