Going Beyond Static Regions of Interest When Assessing APAMs: The Need For A Nuanced Approach (Dynamic Boundaries)

Introduction: Clinicians use alternating-pressure air mattresses (APAMs) to prevent pressure injury (PI) by cyclically offloading regions of pressure. Traditional APAM evaluation methods, using laser Doppler or speckle imaging, rely on single-point or static geographic-based regions of interest (ROI) averages that fail to reflect the non-uniform nature of APAMs. Their effectiveness is difficult to quantify, often leading to misleading results. This study proposes a more nuanced analysis approach beyond geographic ROIs that allows the imaging data to determine the dynamic boundaries (DB).Methods:29 healthy participants (12 male and 17 female) were studied on small-cell APAM surfaces at 2-hour trials with baseline and post-data collection. Multimodal imaging captured 116 x 150-pixel images of oxygenated/deoxygenated hemoglobin, blood flux, and color. Three traditional 30x30-pixel ROIs were chosen: sacrum (higher PI incidence), left glute, and right glute. Comparatively, a global fixed flux threshold was also applied to all the pixels in an image to identify flux values representing concentrated redness, which was compared to visual redness and confirmed with an overlay of the color images.Results:                        Figure 1 shows the differences between participants after 2 hours on an APAM. Each ROI captures only a fraction of concentrated red pixels.     Left Glute (ROI1) Sacrum (ROI2) Right Glute (ROI3) Entire Image (DB) Participant 1 754 2 328 3938 Participant 2 55 273 373 2676 Participant 3 114 5 80 2182 Avg (SD) of DB [All 29 participants] 185.3 (230.1) 99.8 (139.2) 170.7 (197.5) 2373.1 (2438.9) Avg Percent of DB 7.8% 4.2% 7.2%     Table 1. Total number of pixels above fixed-flux threshold for the various ROIs highlights the high variability and low representation of overall concentrated red pixels in each of the ROIs.Discussion: Evaluating tissue response, such as redness, is crucial for assessing support surface effectiveness and injury risk. Individual patient anatomy and APAM’s dynamic pressure delivery are factors in injury risk assessment. However, static geographic ROIs are inadequate for understanding how APAMs affect patients, as evident from the high pixel count variability and low representation of concentrated redness outside these predefined regions. This highlights how conventional ROI analysis misses critical aspects of the surface stimulus, necessitating more nuanced approaches like thresholding or clustering to create Dynamic boundaries. Once the pixels of interest are precisely identified, their responses and behavior can be assessed. Accurate assessment of support surfaces could provide clinical benefits, like better prediction of tissue injury risk and personalized APAM settings optimization.References: