{"doi":"10.1111/ijd.17740","title":"Avatar Annotation to Visualize Disease Evolution in a Chronic Graft‐Versus‐Host Disease Model: A Methodology Paper","abstract":"Current methods for tracking skin disease rely on clinician estimation of disease severity and body surface area (BSA) involvement without standardized methodology. We selected chronic graft-versus-host disease (cGVHD) as a representative condition because cutaneous cGVHD is characterized by the longitudinal evolution of three main morphological groups: epidermal, sclerotic, and pigmentary changes [1]. The National Institutes of Health (NIH) Consensus scoring of cutaneous involvement has been shown to correlate with overall survival [2], and better methods of disease tracking are sorely needed for this potentially fatal disease [3]. Characterization of cutaneous cGVHD is challenging, however, due to the varied morphology, time-consuming nature of scoring, and expertise needed for accurate assessment and documentation of cutaneous involvement. In this methodology study, we explore the utility of annotating findings from clinical photographs employing standardized anatomy mapping software [4, 5] to provide morphology-specific semi-quantitative assessments of disease activity in cutaneous cGVHD. All available photographs from patients with cutaneous cGVHD who provided research consent were reviewed. Five of the most comprehensive photography sets per patient spaced over time were selected for analysis. Oral and genital photographs were excluded due to limited availability. A blinded researcher (F.X.) reviewed all photographic images, with quality checks by a dermatologist with GVHD experience (J.S.L.). For each visit, the extent of epidermal, sclerotic, and pigmentary morphologies was annotated on customized anatomic mapping software to generate standardized research-ready data (Anatomy Mapper) (Figure 1; further data can be found at https://edu.anatomymapper.com/bsa). Avatars were painted with single or combination morphologies on standardized anatomic sites, which generated a tracking matrix representing presence (1)/absence (0)/not assessed (−) for each morphology type, along with visualization of each site. Standardized surface anatomy descriptions and morphology matrix data were generated and collated into a spreadsheet. Spreadsheet data was normalized into laterality and standardized description language pairs, and morphology matrix sums were calculated for each morphology type for each language pair. Language pairs were estimated to BSA percentage based on the previously validated adult Lund-Browder chart. Adjusted BSA was then calculated as a percentage to account for missing photographs. Data manipulation and plots were performed using RStudio Version 4.1.3 (RStudio, PBC, USA). All five patients had Fitzpatrick I–II skin. Adjusted BSA for epidermal, sclerotic, or pigmentary disease varied across five time points and each patient. A representative plot based on the anterior view of avatars is shown in Figure 2. The presenting dermatologic morphology for four patients was epidermal disease. Of these patients, three then progressed to develop sclerotic and pigmentary disease, while one remained with epidermal alongside sclerotic and pigmentary disease. Another developed all three morphologies synchronously. Sclerosis did not consistently develop in areas previously affected by erythema. Isomorphic sclerotic GVHD was noted at the beltline in four patients. The described methods demonstrate the use of standardized avatars to semi-quantitatively analyze clinical photographs as a means to study changes in complex morphologies over time quantitatively. We chose to study patients with cGVHD due to the complex and evolving morphologies and the clinical need for accurate documentation of cutaneous findings in this patient population. However, the principle could be applied to other skin conditions where comparing findings over time or studying patterns of disease distribution or progression across patients is important. Limitations include a small sample size, underrepresentation of darker skin types in our dataset, and the availab","journal":"International Journal of Dermatology","year":2025,"id":553381,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":1,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9596,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2025-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1450404,"name":"Matthew Molenda","orcid":null,"position":1,"is_corresponding":false},{"id":286643,"name":"Edward W. Cowen","orcid":"0000-0003-1918-4324","position":2,"is_corresponding":false},{"id":908081,"name":"Austin Todd","orcid":"0000-0002-6115-4770","position":3,"is_corresponding":false},{"id":1450013,"name":"Kelly Morgan","orcid":"0000-0003-0363-2460","position":4,"is_corresponding":false},{"id":720991,"name":"Julia S. Lehman","orcid":"0000-0002-7389-3853","position":5,"is_corresponding":false},{"id":402616,"name":"Eric R. Tkaczyk","orcid":"0000-0002-2850-4740","position":6,"is_corresponding":false},{"id":1104028,"name":"Fangyi Xie","orcid":"0000-0001-5524-2750","position":0,"is_corresponding":true}],"reference_count":5,"raw_metadata":null,"created_at":"2026-07-19T02:54:41.819682Z","pmid":"40586235","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}