Toward mitigating pressure injuries: Detecting patient orientation from vertical bed reaction forces

Abstract

Introduction: Prolonged bed rest without repositioning can lead to pressure injuries. However, it can be challenging for caregivers and patients to adhere to repositioning schedules. A device that alerts caregivers when a patient has remained in the same orientation for too long may reduce the incidence and/or severity of pressure injuries. This paper proposes a method to detect a person's orientation in bed using data from load cells placed under the legs of a hospital grade bed. Methods: Twenty able-bodied individuals were positioned into one of three orientations (supine, left side-lying, or right side-lying) either with no support, a pillow, or a wedge, and the head of the bed either raised or lowered. Breathing pattern characteristics extracted from force data were used to train two machine learning classification systems (Logistic Regression and Feed Forward Neural Network) and then evaluate for their ability to identify each participant's orientation using a leave-one-participant-out cross-validation. Results: The Feed Forward Neural Network yielded the highest orientation prediction accuracy at 94.2%. Conclusions: The high accuracy of this non-invasive system's ability to a participant's position in bed shows potential for this algorithm to be useful in developing a pressure injury prevention tool.

Detecting Patient Position Using Bed-Reaction Forces and Monitoring Skin-Bed Interface Forces for Pressure Injury Prevention and Management

“…The instrumentation was set up in a similar manner to Wong et al (12). Data was collected in CareLab, a simulated patient care environment located within Toronto Rehabilitation Institute (TRI), using a Carroll hospital bed (Carroll Hospital Group, Kalamazoo, MI).…”

“…The proof-of-concept work was able to detect healthy participant position (supine, left-side, or right-side) with 94.2% accuracy (n=20). When Wong et al’s (12) model was tested on the data collected from nine older adults sleeping in their own beds at home, the accuracy dropped to ∼88.5%. The drop in accuracy was suspected to be due to the large variations of sleeping positions that can be adopted.…”

Measuring Repositioning in Home Care for Pressure Injury Prevention and Management

“…The filtfilt function was used to bandpass filter the sum of the LH and RH signals ensuring a zero-phase shift. Our data was processed similar to Wong et al [18]. Each data point used for training/testing our machine learning classifiers was the average of a 45 s moving window that contained 2250 observations, where a new value was computed by shifting the window by 15 s. Ground Truth Participant Position Labels -Three members of the research team independently reviewed the time-lapse images for each data point and assigned a label indicating the participant was in one of three positions: right-side-lying, left-side-lying, or supine for each 45 s. The most common label from the three reviewers was selected as the final ground truth label.…”

“…This validation approach was selected to maximize the number of training observations that were used for each classifier. Incremental Learning -An incremental learning approach was used to evaluate the ability of the classifier to adapt to the data from the left-out participant in a similar manner as described in Wong et al [18]. The machine learning classifier was iteratively trained using different percentages of the left-out participant's data (c, with c = 0%, 10%, 20%, or 30%).…”

“…Sensors collect force data and features are extracted from the data. Using advanced machine learning algorithms, we have been able to predict an individual’s position in bed as either supine, right-side-lying or left-side-lying with 94.2% accuracy in an controlled laboratory environment [18]. Our ultimate goal is to build on this position detection system to develop a repositioning prompting system that will continuously monitor an patient’s position in bed and alert a caregiver when the patient has remained in the same position for too long (e.g.…”