No abstract

Accurate and reproducible color capture is vital in medical photography. Camera distance and angle are particularly important as they are highly variable in a clinical setting. To account for variability in illumination, camera technology, and geometric effects, color standards are often used for color correction. To explore how geometry affects color, we quantified the change in CIELAB color value of a color standard for diverse skin tones at varying smartphone camera distances and angles. Whereas both chromaticity (a* and b*) and lightness (L*) were affected by angle, distance only affected L* (standard error of measurement, SEM > 1 CIELAB unit). Flash usage did not generally reduce distance and angle associated variability. Compared to compressed (JPG) format, raw (DNG) images had decreased median variability across different distances and angles. These findings suggest that in medical photography, inconsistent camera distance and angle can increase variability in photographed skin appearance over time.

Cutaneous erythema is used in diagnosis and response assessment of cutaneous chronic graft-versus-host disease (cGVHD). The development of objective erythema evaluation methods remains a challenge. We used a pre-trained neural network to segment cGVHD erythema by detecting changes relative to a patient's registered baseline photo. We fixed this change detection algorithm on human annotations from a single photo pair, by using either a traditional approach or by marking definitely affected ("Do Not Miss", DNM) and definitely unaffected skin ("Do Not Include", DNI). The fixed algorithm was applied to each of the remaining 47 test photo pairs from six follow-up sessions of one patient. We used both the Dice index and the opinion of two board-certified dermatologists to evaluate the algorithm performance. The change detection algorithm correctly assigned 80% of the pixels, regardless of whether it was fixed on traditional (median accuracy: 0.77, interquartile range 0.62-0.87) or DNM/DNI segmentations (0.81, 0.65-0.89). When the algorithm was fixed on markings by different annotators, the DNM/ DNI achieved more consistent outputs (median Dice indices: 0.94-0.96) than the traditional method (0.73-0.81). Compared to viewing only rash photos, the addition of baseline photos improved the reliability of dermatologists' scoring. The inter-rater intraclass correlation coefficient increased from 0.19 (95% confidence interval lower bound: 0.06) to 0.51 (lower bound: 0.35). In conclusion, a change detection algorithm accurately assigned erythema in longitudinal photos of cGVHD. The reliability was significantly improved by exclusively using confident human segmentations to fix the algorithm. Baseline photos improved the agreement among two dermatologists in assessing algorithm performance.