Baseline Photos and Confident Annotation Improve Automated Detection of Cutaneous Graft-Versus-Host Disease

Xiaoqi Liu; Kelsey Parks; Inga Saknite; Tahsin Reasat; Austin D Cronin; Lee E Wheless; Benoit M Dawant; Eric R Tkaczyk

doi:10.2991/chi.k.210704.001

Baseline Photos and Confident Annotation Improve Automated Detection of Cutaneous Graft-Versus-Host Disease

Clin Hematol Int. 2021 Jul 15;3(3):108-115. doi: 10.2991/chi.k.210704.001. eCollection 2021 Sep.

Authors

Xiaoqi Liu^{1

2

3}, Kelsey Parks^{1

3}, Inga Saknite³, Tahsin Reasat², Austin D Cronin^{1

3}, Lee E Wheless^{1

3}, Benoit M Dawant², Eric R Tkaczyk^{1

3

4}

Affiliations

¹ Department of Veterans Affairs, Tennessee Valley Healthcare System, Dermatology Service, 1310 24th Avenue South, Nashville, TN 37212-2637, USA.
² Department of Electrical Engineering and Computer Science, Vanderbilt University, 361 Jacobs Hall, Nashville, TN 37235-1662, USA.
³ Vanderbilt Dermatology Translational Research Clinic (VDTRC.org), Department of Dermatology, Vanderbilt University Medical Center, 719 Thompson Lane, One Hundred Oaks Suite 26300, Nashville, TN 37204, USA.
⁴ Department of Biomedical Engineering, Vanderbilt University, Nashville, TN, USA.

Abstract

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.

Keywords: Cutaneous erythema; baseline photos; change detection; chronic graft-versus-host disease; computer vision; convolutional neural networks; longitudinal photos.

Abstract

Grants and funding