Testing the validity of erythema detection algorithms.

Area of Science:

• Dermatology
• Biomedical Optics
• Medical Imaging

Background:

• Visual inspection is standard for erythema detection but is limited in darkly pigmented skin due to melanin masking.
• Tissue Reflectance Spectroscopy (TRS) is a noninvasive technique for quantifying skin color and detecting erythema.
• Recent applications of TRS include characterizing erythema from reactive hyperemia and Stage I pressure ulcers.

Purpose of the Study:

• To compare the reliability and validity of different erythema detection algorithms using Tissue Reflectance Spectroscopy.
• To determine the sensitivity and specificity of these algorithms.
• To assess algorithm performance across varying skin pigmentation levels.

Main Methods:

• Evaluation of erythema detection algorithms using Tissue Reflectance Spectroscopy (TRS).
• Calculation of sensitivity and specificity for each algorithm.
• Comparison of algorithm validity across subjects with different skin pigmentation.

Main Results:

• Two algorithms, Diffey and Helen Hayes Hospital (HHH), demonstrated high sensitivity (>85%) and specificity (>75%).
• Most evaluated algorithms showed adequate validity across all subjects.
• The HHH algorithm's validity remained consistent regardless of the subject's skin pigmentation.

Conclusions:

• TRS is a viable tool for erythema detection, particularly beneficial for individuals with darker skin tones.
• The HHH algorithm is a robust and reliable option for erythema detection across diverse skin pigmentation.
• These findings support the use of TRS in research involving erythema assessment in varied populations.