Objectives: The assessment of systemic sclerosis–associated Raynaud phenomenon (ASRAP) questionnaire was recently developed. We tested the convergent validity of the ASRAP instrument with patient demographics and capillary morphology on nailfold capillaroscopy (NFC). Methods: Participants completed the ASRAP (range 20–80; higher scores unfavourable) and underwent NFC using Dinolite™. We collected demographic and relevant clinical data. Two blinded assessors classified NFC images as ‘scleroderma pattern’ present or absent and recorded specific abnormalities. We assigned the NFC semi-quantitative score (range 0–3) based on the frequency of these abnormalities. Descriptive statistics were generated, and regression and correlation analyses were used to evaluate associations between ASRAP, NFC and clinical factors. Results: Seventy-five patients [87% women, 78% white, 77% with limited cutaneous SSc] with a mean (SD) age of 58 (14) were recruited from scleroderma clinics. ASRAP scores were 11.2 points higher in women compared with men with other significant variables held constant. The NFC semi-quantitative score weakly correlated with ASRAP (Spearman’s σ = 0.23, P = 0.05), with moderate correlation in winter (Spearman’s σ = 0.36, P = 0.04). Holding other variables constant, patients with dilated capillaries and capillary dropout had higher ASRAP scores by 6.1 and 6.8 points, respectively. Conclusion: In SSc patients, females reported higher ASRAP scores. Greater proportion of ‘scleroderma pattern’ NFC abnormalities as well as specific abnormalities like dilated capillaries and capillary dropout were associated with higher ASRAP providing mechanistic insight and supporting the construct validity of ASRAP for assessing SSc-RP.
As interest grows in multimodal large language models (LLMs) for ophthalmic imaging, their reliability must be established before clinical use. A key prerequisite is determining whether image quality permits interpretation. However, their ability to assess retinal nerve fiber layer (RNFL) optical coherence tomography (OCT) scan quality remains uncertain. We compared a general-purpose AI tool (ChatGPT-5.1) with a clinical AI tool (DoxGPT) for RNFL OCT report-quality assessment. Both evaluated 145 bilateral reports from 11 patients in an enriched quality-control sample using global and stepwise prompts. One ophthalmologist’s responses to the same predefined quality questions served as the clinical reference for calculating percent agreement, linearly weighted Cohen’s κ with patient-level cluster-bootstrap 95% confidence intervals (CIs), and prevalence-adjusted bias-adjusted κ (PABAK). Global prompting showed higher observed agreement than stepwise prompting: ChatGPT, 56% versus 47% (κ = 0.36 [95% CI, 0.23–0.49] versus 0.12 [0.08–0.18]); DoxGPT, 59% versus 48% (κ = 0.41 [0.28–0.54] versus 0.13 [0.03–0.23]). Component agreement was higher for signal strength (ChatGPT/DoxGPT: 99%/99%) and optic nerve head centering (87%/96%), and lower for dropout (41%/43%) and segmentation error (64%/65%). Both tools showed limited agreement with the ophthalmologist for overall RNFL report reliability. Their accuracy requires independent validation, supporting continued clinician review.