Longitudinal In-Bed Pressure Signals Decomposition and Gradients Analysis for Pressure Injury Monitoring

Abstract

Pressure injury (PI) is a major problem for patients that are bound to a wheelchair or bed, such as seniors or people with spinal cord injuries. This condition can be life threatening in its later stages. It can be very costly to the healthcare system as well. Fortunately with proper monitoring and assessment, PI development can be prevented. The major factor that causes PI is prolonged interface pressure between the body and the support surface. A possible solution to reduce the chance of developing PI is changing the patient’s in-bed pose at appropriate times. Monitoring in-bed pressure can help healthcare providers to locate high-pressure areas, and remove or minimize pressure on those regions. The current clinical method of interface pressure monitoring is limited by periodic snapshot assessments, without longitudinal measurements and analysis. In this paper we propose a pressure signal analysis pipeline to automatically eliminate external artefacts from pressure data, estimate a person’s pose, and locate and track high-risk regions over time so that necessary attention can be provided.

Optimal Image Characterization for In-Bed Posture Classification by Using SVM Algorithm

“…The ROC curve closest to perfect classification is the one for the posture P 4 , with 83.3%, in Figure 9b. In spite of the classification accuracy being less than that reported in [23], it is worth highlighting that the principal purpose of this work was to simplify the stage of image pre-processing. The results obtained employing the proposed methodology are considered excellent based on established metrics.…”

“…The proposed methodology was applied to a structured database with a total of 208 images of thirteen different patients in sixteen bed positions. In a previous work [23], three positions were identified; however, the pressure images showed different pressure zones.…”

“…In particular, the pressure images have a low resolution, and external objects hide the relevant information for ML algorithms. Given the diversity of variables, several strategies have been proposed to classify the postures based on image processing [23][24][25][26]. In these works, a pre-processing stage is suggested that refers to the conditioning of the images before they are processed by the ML algorithm.…”