Impact of Imaging Protocols on Thermal Detection of Pressure Injuries: Threshold versus Deep Learning Across Skin Tones

Authors

Abstract

Pressure injuries represent a significant healthcare challenge requiring early detection to prevent severe complications. While thermal imaging shows promise for detecting early pressure related temperature changes, its robustness across varying imaging conditions and diverse patient populations remains unclear. This study systematically evaluated how imaging protocol variations (lighting, distance, positioning, camera type) and participant skin tone influence classification model performance for thermal cooling detection. Using a controlled cooling protocol to simulate early pressure injury temperature changes, we collected 1,680 images from 35 diverse participants across 12 imaging protocol variations. We compared two approaches: three deep learning models (MobileNetV2, InceptionNetV3, ResNet50) and a threshold-based approach using an optimal fixed threshold temperature differential. Deep learning models outperformed the threshold-based approach, achieving 98.699.6% accuracy compared to 95.6%, with superior performance across all imaging protocols and skin tone groups. Threshold-based approach showed camera-dependent misclassification patterns across skin tones. On the high-resolution FLIR E8XT, the MST 7-10 group had 8 of 11 misclassifications. This pattern shifted on the low-resolution FLIR ONE Pro, where the intermediate skin tone group (MST 6) had 22 of 44 total misclassifications. In contrast, deep learning models maintained consistent performance across all skin tone groups and imaging protocols. Visualization analysis of the deep learning models suggested that these models focused on thermal gradients at cooling region boundaries, suggesting that spatial temperature gradients, not single-value thresholds, are critical for accurate detection. These findings suggest the potential of deep learning-based approaches to maintain robust, equitable performance across diverse skin tones and imaging conditions. Author SummaryPressure injuries are a major clinical challenge requiring early detection. Current visual inspection methods are unreliable, especially for patients with darker skin tones. Thermal imaging shows promise for detecting early temperature changes, but no studies have systematically evaluated how imaging variations affect detection accuracy across diverse populations. Using two cameras, FLIR E8XT (320x240 pixels) and FLIR ONE Pro (160x120 pixels), we collected 1,680 thermal images from 35 healthy adults across diverse skin tones within a controlled setting with simulated cooling and evaluated how imaging variations (lighting, distance, positioning, camera type) affect performance on two classification approaches: deep learning models that preserve spatial temporal context, and threshold-based approach using a single fixed temperature. Deep learning based approaches demonstrated superior robustness to camera type compared to the threshold based approach which exhibited hardware dependency, with performance deteriorating dramatically on the lower resolution FLIR ONE Pro camera. Deep learning-based approaches also showed consistent performance across all skin tone groups, indicating that both camera selection and labeling methodology are critical for clinical thermal imaging systems. Deep learning models preserve spatial temperature information and show promise for reliable pressure injury detection across diverse patient populations.

Tags

Topics