Effects of epidermal pigmentation on the accuracy of hyperspectral versus commercial near-infrared spectroscopy tissue oximeters
Bibliographic record
Abstract
The accuracy of pulse and tissue oximeters has been questioned due to potential bias related to skin pigmentation. Current commercial oximeters use a limited number of wavelengths to estimate tissue blood oxygen saturation (StO2), failing to account for additional light absorption caused by melanin which particularly important in individuals with darker skin. This limitation has resulted in inaccurate readings, missed diagnoses, and delayed treatment for these people. Interestingly, the absorption spectrum of melanin in the near-infrared region decreases quasi-monotonically with wavelength, similar to light scattering, which can be effectively mitigated by applying spectral derivatives to hyperspectral near-infrared spectroscopy (h-NIRS) measurements. Consequently, we hypothesize that h-NIRS can mitigate the confounding effects of melanin, to provide more accurate measurements of StO2 across diverse skin tones. This study evaluates the accuracy of h-NIRS in comparison to few-wavelength spectroscopy (i.e. simulated commercial tissue oximeters) through simulations and tissue-mimicking phantom experiments. The accuracy of h-NIRS for determining StO2 was assessed against two “virtual devices”, each mimicking a commercially available tissue oximeter (INVOS-5100c and NIRO- 200NX). Results showed significant skin tone-dependent biases of the simulated commercial devices, while h-NIRS demonstrated consistently low errors (<3%) across all skin tones and oxygenation levels. These findings confirm that h-NIRS can effectively mitigate skin tone biases, achieving superior accuracy compared to these commercial devices. These findings are significant as they offer a viable solution to a major limitation of current tissue oximeters and establish h-NIRS as a promising technique for inclusive assessment of StO2.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.012 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".