Skip to content
Decision-ready map
• Specify confounders as requirements: hair, tattoos, scars, cosmetics, dyes
• Engineer detection: optical SNR + contact/pressure + ambient-light rejection
• Validate worst-case: dark polish, pigment inks, scar scattering changes
• Label clearly: contraindications, alternative sites, expected failure modes
• Align with standards (ISO 80601-2-61) and post-update revalidation
(1) What it is
Theme 6 is a design-and-labeling problem: hair, tattoos, scars, and cosmetics are predictable optical perturbations that sit at the sensor–tissue interface. They can reduce coupling, add pigment absorption, and change scattering, pushing your system into low-SNR regimes or biasing spectral baselines. Because many products convert optical signals into thresholded decisions (alerts, scores, classifications), small confounder-driven shifts can become clinically meaningful. ‘Technology design and labeling’ means you must: (a) design for robustness and safe failure, (b) validate interference systematically, and (c) communicate limits clearly in labeling and UI so users can respond correctly.
(2) Who it helps
This brief helps founders defining intended use and claims, engineering requirements, clinical validation protocols, user instructions, and postmarket monitoring. It applies to consumer wearables that may drift toward medical use, as well as regulated devices in hospitals and remote monitoring programs.
(3) What evidence exists
Interference evidence exists and is actionable for product requirements. Nail polish systematically affects pulse oximetry readings at small magnitude in many healthy-subject studies and has been synthesized in a 2023 systematic review; some colors can also cause measurement failure. Surgical dyes (patent blue V and related dyes) can produce factitious desaturation and interfere with pulse oximetry/co-oximetry—demonstrating that exogenous pigments in the optical path can break simple ratio assumptions. Tattoo inks contain diverse pigments; micro-Raman studies identify in situ ink composition, and phantom-based Raman work shows how scattering matrices affect spectra—highlighting why tattooed skin is not a neutral background for spectroscopy. Scar tissue alters collagen/water and scattering/oxygenation estimates in DRS studies of keloids, implying that inverse-model priors derived from ‘normal skin’ are not generalizable. Cosmetics measurably change surface reflectance and absorbance; optical methods can estimate foundation-layer properties, and reflectance spectroscopy is explicitly used to quantify skin response to topical products—so your device must either control for cosmetics or detect and flag their presence.
(4) Translation barriers
Common translation barriers include: (1) failure to scope ‘interferents’ early, leading to late redesign; (2) lack of an interference test matrix (colors, pigments, scar types, hair density, application thickness); (3) overreliance on average accuracy instead of worst-case and ‘measurement failure’ rates; (4) insufficient human factors work (users do not read long IFUs; they need UI prompts and clear warnings); and (5) update drift—firmware/ML updates can change robustness unless change control triggers revalidation.
(5) Equity/safety checks
Confounders overlap with equity: tattoos and cosmetics are more prevalent in some communities and occupations; scars can be more common after injury or surgery; hair density varies widely. A product that ‘just fails’ in these conditions can exclude users or create uneven safety. Build ‘safe failure’ behavior: detect low-quality signals, suppress unreliable outputs, and provide an alternative-site instruction. Ensure labeling is non-stigmatizing and includes practical workarounds rather than prohibitions.
(6) Decision questions
• What are the known interferents for our modality (polish/gel, dyes, tattoo pigments, scar tissue, hair/coupling) and are they in our requirements doc?
• Do we measure and log quality (SNR, perfusion index, contact pressure, ambient light) and use it to gate outputs?
• Do our validation datasets include confounders at realistic prevalence and worst-case conditions?
• Does labeling/UI instruct users what to do (remove cosmetics, relocate site) and when to seek confirmatory assessment?
• What is our revalidation trigger after hardware/firmware/ML updates, and how do we monitor confounder-related field complaints?
(7) Practical next steps
1) Build an interference matrix: nail cosmetics (colors/types), tattoo pigments, scar types (keloid/burn), hair density, topical products; define pass/fail metrics including ‘unable to measure.’
2) Engineer robust coupling and sensing: multi-wavelength options, ambient-light rejection, pressure/contact sensing, and algorithmic quality gating.
3) Validate worst-case and publish stratified results; include both bias and failure-rate metrics.
4) Implement labeling + UI prompts: ‘If tattoo/polish/scar/hair at site → relocate’ and include alternative sites.
5) Add postmarket monitoring: track confounder-linked failure reports; revalidate after updates per QMS and standards expectations.
(8) References
Aggarwal AN, Agarwal R, Dhooria S, et al. Impact of Fingernail Polish on Pulse Oximetry Measurements: A Systematic Review. Respiratory Care. 2023.
https://doi.org/10.4187/respcare.10399
Yeganehkhah M, Dadkhahtehrani T, Bagheri AR, Kachoie A. Effect of Glittered Nail Polish on Pulse Oximetry Measurements in Healthy Subjects. Iran J Nurs Midwifery Res. 2019.
https://doi.org/10.4103/ijnmr.IJNMR_176_17
Hueter L, Schwarzkopf K, Karzai W. Interference of patent blue V dye with pulse oximetry and co-oximetry. Eur J Anaesthesiol. 2005.
https://doi.org/10.1017/S0265021505230818
Howard JD, Moo V, Sivalingam P. Anaphylaxis and other adverse reactions to blue dyes: a case series. Anaesth Intensive Care. 2011.
https://doi.org/10.1177/0310057X1103900221
Piñero A, Illana J, García-Palenciano C, et al. Effect on Oximetry of Dyes Used for Sentinel Lymph Node Biopsy. Arch Surg. 2004.
https://doi.org/10.1001/archsurg.139.11.1204
Poon KWC, Dadour IR, McKinley AJ. In situ chemical analysis of modern organic tattooing inks by micro-Raman spectroscopy. J Raman Spectrosc. 2008.
https://doi.org/10.1002/jrs.1973
Sadura F, Wróbel MS, Karpienko K. Colored Tattoo Ink Screening Method with Optical Tissue Phantoms and Raman Spectroscopy. Materials (Basel). 2021.
https://doi.org/10.3390/ma14123147
Hsu C-K, Tzeng S-Y, Yang C-C, et al. Non-invasive evaluation of therapeutic response in keloid scar using diffuse reflectance spectroscopy. Biomed Opt Express. 2015.
https://doi.org/10.1364/BOE.6.000390
Tseng S-H, Hsu C-K, Lee JY-Y, et al. Noninvasive evaluation of collagen and hemoglobin in keloid scars using DRS. J Biomed Opt. 2012.
https://doi.org/10.1117/1.JBO.17.7.077005
Yoshida K, Okiyama N. Estimation of reflectance/transmittance/absorbance of cosmetic foundation layer on skin. Opt Express. 2021.
https://doi.org/10.1364/oe.442219
Mancuso A, d’Avanzo ND, Cristiano MC, Paolino D. Reflectance spectroscopy to explore skin reactions to topical products. Front Chem. 2024.
https://doi.org/10.3389/fchem.2024.1422616
Kim KB, Baek HJ. Photoplethysmography in Wearable Devices: A Comprehensive Review. Electronics. 2023.
https://doi.org/10.3390/electronics12132923
Cooksey CC, Allen DW, Tsai BK. Reference Data Set of Human Skin Reflectance. J Res Natl Inst Stan. 2017.
https://doi.org/10.6028/jres.122.026
Cooksey CC, Allen DW, Tsai BK. Reference Data Set of Human Skin Reflectance (data). NIST. 2017.
https://doi.org/10.18434/M38597
IEC. ISO 80601-2-61:2026 Pulse oximeter equipment — safety and essential performance.
https://webstore.iec.ch/en/publication/74527