University of North Carolina studies effects of skin pigmentation on oximetry

22 Jul 2025

Details of optical interactions help explain observed bias in real-world results.

A project led by the University of North Carolina at Chapel Hill (UNC) has studied the effects of skin pigmentation on measurements of blood oxygenation.

The extent to which different skin types influence objective measurements of oxygenation by pulse oximetry techniques has become a topic of significant interest for clinicians.

Among recent studies of the problem, a Brown University project developing a smartphone-based oximetry platform was motivated partly by the need to standardize skin tone assessment in pulse oximetry. A UK report in 2024 urged makers of pulse oximeters to design better and smarter devices that are not biased by skin tone.

The UNC study, published in Biophotonics Discovery, aimed to experimentally characterize the impact of skin pigmentation on pulse oximeters across a broad range of oxygen saturations, and determine whether skin pigmentation can indeed explain the observed bias in results for patients of color.

"Within the medical community it is well-known that there are elevated rates of ICU mortality in racial minorities," noted the project in its paper. "It is possible that a technological gap, namely the inaccuracy of pulse oximeter data for racial and ethnic minorities, may be contributing to this mortality gap."

The researchers in the UNC lab of Wubin Bai used Hampshire pigs as an animal model, exploiting the fact that the animals naturally have both dark and light patches of skin. This means the team could compare measurements from pigmented and nonpigmented areas on the same animal under identical conditions.

Seeing red: the key light-tissue interaction

In trials the project focused on the spectral characteristics of light interacting with four key chromophores in skin: melanin, melanosomes, deoxyhemoglobin, and oxyhemoglobin.

Because the first two are responsible for skin pigmentation differences and the latter two account for changes in oxygen saturation, analysis of these four provides a basis to understand how skin pigmentation interacts with oxygen saturation in pulse oximetry, commented the team.

The team built custom pulse oximeters and applied them to areas of both skin types on the animal subjects. It gradually reduced the pigs’ oxygen levels from 100 percent to 70 percent by adjusting the mix of oxygen and nitrogen in the air, while collecting arterial blood samples for separate laboratory assessment of oxygen saturation as a comparison.

Signal processing analysis let the project separate the pulsatile (heartbeat-related) and baseline components of the infrared light signals captured by the pulse oximeters, and the changes in signal with oxygen levels between pigmented and nonpigmented skin were calculated.

Results revealed fine details of the skin's effect on pulse oximetry, including how the red pulsatile index is significantly more attenuated in pigmented skin, especially in hypoxia, whereas the near-IR pulsatile index sensitivity is unaffected. This effect reduces the device's ability to detect changes in blood oxygen levels, particularly during low-oxygen conditions where accurate readings are most critical.

"Skin pigmentation attenuates the sensitivity of red-light absorption to changes in oxygen saturation, and pulse oximeter output is consequently attenuated in pigmented skin," said the project. "Future work should focus on how light-tissue interactions in the red region are uniquely affected by skin pigmentation."