ChatGPT as a New Tool to Select a Biological for Chronic Rhino Sinusitis with Polyps, “Caution Advised” or “Distant Reality”?

Abstract

ChatGPT is an advanced language model developed by OpenAI, designed for naturallanguage understanding and generation. It employs deep learning technology to comprehend andgenerate human-like text, making it versatile for various applications. The aim of this study is toassess the alignment between the Rhinology Board’s indications and ChatGPT’s recommendationsfor treating patients with chronic rhinosinusitis with nasal polyps (CRSwNP) using biologic ther-apy. An observational cohort study involving 72 patients was conducted to evaluate various param-eters of type 2 inflammation and assess the concordance in therapy choices between ChatGPT andthe Rhinology Board. The observed results highlight the potential of Chat-GPT in guiding optimalbiological therapy selection, with a concordance percentage = 68% and a Kappa coefficient = 0.69(CI95% [0.50; 0.75]). In particular, the concordance was, respectively, 79.6% for dupilumab, 20% formepolizumab, and 0% for omalizumab. This research represents a significant advancement in man-aging CRSwNP, addressing a condition lacking robust biomarkers. It provides valuable insightsinto the potential of AI, specifically ChatGPT, to assist otolaryngologists in determining the optimalbiological therapy for personalized patient care. Our results demonstrate the need to implement theuse of this tool to effectively aid clinicians.