Fitzpatrick Thresholding for Skin Image Segmentation

doi:10.1007/978-3-032-05825-6_2

Fitzpatrick Thresholding for Skin Image Segmentation

2026-01-01

0

PRE

AI

D

Duncan Bayard Stothers

DOI：10.1007/978-3-032-05825-6_2

Original

Original request for help

Save

Abstract

En 中文

Accurate estimation of the body surface area (BSA) involved by a rash, such as psoriasis, is critical for assessing rash severity, selecting an initial treatment regimen, and following clinical treatment response. Attempts at segmentation of inflammatory skin disease such as psoriasis perform markedly worse on darker skin tones, potentially impeding equitable care. We assembled a psoriasis dataset sourced from six public atlases, annotated for Fitzpatrick skin type, and added detailed segmentation masks for every image. Reference models based on U-Net, ResUNet, and SETR-small are trained without tone information.On the tuning split, we sweep decision thresholds and select (i) global optima and (ii) per-Fitzpatrick skin tone optima for Dice and binary IoU. Adapting Fitzpatrick-specific thresholds lifted segmentation performance for the darkest subgroup (Fitz VI) by up to +31% bIoU and +24% Dice on U-Net, with consistent, though smaller, gains in the same direction for ResUNet (+25% bIoU, +18% Dice) and SETR-small (+17% bIoU, +11% Dice).Because Fitzpatrick skin tone classifiers trained on Fitzpatrick17 now exceed 95% accuracy, the cost of skin tone labeling required for this technique has fallen dramatically. Fitzpatrick thresholding is simple, model-agnostic, requires no architectural changes, no re-training, and is virtually cost-free. We demonstrate the inclusion of Fitzpatrick thresholding as a potential future fairness baseline.

Keywords:

Fitzpatrick

Psoriasis

Segmentation

BSA

Fitzpatrick Thresholding for Skin Image Segmentation

Abstract

Journal

Organization