Comparison of artificial intelligence and rheumatologists in nailfold capillaroscopic evaluation.

OBJECTIVE

The aim of this study was to evaluate the diagnostic accuracy of a pretrained Vision Transformer model on nailfold capillaroscopy images, comparing its performance to that of expert rheumatologists.

METHODS

We retrospectively analyzed 104 anonymized images (23 normal, 81 pathological) from publicly available datasets. The pretrained Vision Transformer model was applied without fine-tuning. Two rheumatologists independently assessed the same image set. Accuracy and interrater agreement were calculated.

RESULTS

The AI model produced 0% clinical applicability, failing to generate meaningful classifications. In contrast, the consensus of two rheumatologists achieved the highest diagnostic performance, with 94.2% accuracy and a Cohen's Kappa of 0.827. Individual rheumatologist evaluations yielded comparatively lower accuracy and agreement.

CONCLUSION

We retrospectively analyzed X anonymized images (XX normal, XX pathological) from publicly available datasets. The pretrained Vision Transformer model was applied without fine-tuning. Two rheumatologists independently assessed the same image set. Accuracy and interrater agreement were calculated.