Thermal Imaging for Pressure Injury Detection Across Skin Tones

This direction is merged into another direction: melanin-induced signal attenuation. Its last version stays readable and citable here.

Version 1 of 17 papers19 conceptsfrozen 2026-10-04

Traditional clinical assessment for early-stage pressure injuries (PrIs) relies heavily on identifying erythema—the localized redness of the skin caused by increased blood flow. However, this visual signal is notoriously unreliable in patients with darker skin tones, where melanin masks the subtle color changes of tissue stress.

Recent research by Sharon Eve Sonenblum, Kathleen S. Jordan, and colleagues suggests a shift from the visible spectrum to the infrared. By measuring emitted heat rather than reflected light, thermal imaging provides a more objective, physiologically grounded window into tissue health that persists regardless of skin pigmentation.

The Decoupling of Redness and Heat

The fundamental problem with visual inspection is that redness and temperature are not perfectly correlated. In Impact of skin tone and cupping on erythema and thermal imaging measurements, Kathleen S. Jordan and Vicki Hertzberg demonstrate that visual erythema and skin temperature track different underlying processes.

While redness indicates blood accumulation (hyperemia), temperature changes reflect active perfusion and metabolic shifts. Because these signals are "decoupled," relying solely on visual cues provides an incomplete picture. In darker skin, where the visual signal is muted, the thermal signal remains a robust indicator of the hemodynamic activity associated with early tissue damage.

Emitted Radiation vs. Reflected Light

To understand why thermal imaging is more equitable, we must look at the physics of the signal. Optical photography depends on reflected light. A camera captures photons that bounce off the skin's surface; because melanin absorbs these photons, it fundamentally changes the data the camera receives.

In contrast, long-wave infrared thermography detects emitted radiation. As Miriam Asare-Baiden explains in A Feasibility Study of Thermography for Detecting Pressure Injuries Across Diverse Skin Tones, the body acts as a heat source. The infrared energy captured by the sensor is a product of the tissue's internal state rather than external illumination. This makes the modality inherently more "demographically neutral" because the physiological indicator—heat—is visible even when the visual indicator—redness—is not.

The Melanin Interaction

While thermal imaging is more consistent than visual inspection, it is not entirely immune to the effects of skin tone. A recurring finding in work by Sharon Eve Sonenblum and Glory Tomi John is that higher melanin concentrations correlate with more pronounced measured temperature changes during thermal stress (Impact of Skin Tone, Environmental, and Technical Factors on Thermal Imaging).

There are two primary hypotheses for why this happens:

  1. Emissivity: Melanin might slightly alter the skin's ability to emit infrared radiation.
  2. Thermodynamics: Darker skin might have different heat dissipation properties or physiological responses to stress.

Because of this interaction, Sonenblum argues that clinical "thresholds" for what constitutes a dangerous temperature rise cannot be one-size-fits-all. Instead, diagnostic algorithms must incorporate melanin-specific thresholds to ensure accuracy across all populations.

Robustness in Clinical Environments

Beyond equity, thermal imaging offers technical advantages for automated detection. In Impact of Imaging Protocols on Convolutional Neural Network-Based Pressure Injury Detection, researchers including Judy Wawira Gichoya found that deep learning models trained on thermal data are significantly more robust than those trained on optical photos.

Optical models are sensitive to "noise"—changes in room lighting, shadows, and the distance between the camera and the patient. Thermal models, however, maintain over 90% accuracy regardless of these environmental factors. By focusing on the heat signature, the AI bypasses the visual artifacts that typically plague computer vision in clinical settings, providing a reliable foundation for early, equitable diagnosis.

Papers in this version

History & citationVersion 1 · 2026-10-04

Merged into another direction · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_thermal_imaging_for_pressure_injury_dete_4b36afb6,
  title = {Thermal Imaging for Pressure Injury Detection Across Skin Tones},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/thermal-imaging-for-pressure-injury-detection-across-skin-tones/rdr_4b36afb669cb493d826ed675fce19c4c},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}

This exact version (version 1, frozen text):

@misc{lacuna_thermal_imaging_for_pressure_injury_dete_4b36afb6_v1,
  title = {Thermal Imaging for Pressure Injury Detection Across Skin Tones},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/thermal-imaging-for-pressure-injury-detection-across-skin-tones/rdr_4b36afb669cb493d826ed675fce19c4c/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}