Search

Objective: To investigate the consistency of Chatbot Generative Pretrained Transformer (ChatGPT)-4 in the analysis of clinical pictures of common laryngological conditions., Study Design: Prospective uncontrolled study., Setting: Multicenter study., Methods: Patient history and clinical videolaryngostroboscopic images were presented to ChatGPT-4 for differential diagnoses, management, and treatment(s). ChatGPT-4 responses were assessed by 3 blinded laryngologists with the artificial intelligence performance instrument (AIPI). The complexity of cases and the consistency between practitioners and ChatGPT-4 for interpreting clinical images were evaluated with a 5-point Likert Scale. The intraclass correlation coefficient (ICC) was used to measure the strength of interrater agreement., Results: Forty patients with a mean complexity score of 2.60 ± 1.15. were included. The mean consistency score for ChatGPT-4 image interpretation was 2.46 ± 1.42. ChatGPT-4 perfectly analyzed the clinical images in 6 cases (15%; 5/5), while the consistency between GPT-4 and judges was high in 5 cases (12.5%; 4/5). Judges reported an ICC of 0.965 for the consistency score (P = .001). ChatGPT-4 erroneously documented vocal fold irregularity (mass or lesion), glottic insufficiency, and vocal cord paralysis in 21 (52.5%), 2 (0.05%), and 5 (12.5%) cases, respectively. ChatGPT-4 and practitioners indicated 153 and 63 additional examinations, respectively (P = .001). The ChatGPT-4 primary diagnosis was correct in 20.0% to 25.0% of cases. The clinical image consistency score was significantly associated with the AIPI score (r s  = 0.830; P = .001)., Conclusion: The ChatGPT-4 is more efficient in primary diagnosis, rather than in the image analysis, selecting the most adequate additional examinations and treatments., (© 2024 American Academy of Otolaryngology–Head and Neck Surgery Foundation.)

Objectives: To review the normative data for acid, weakly acid, and nonacid proximal esophageal (PRE) and hypopharyngeal reflux (HRE) events in diagnosing laryngopharyngeal reflux (LPR) using ambulatory reflux monitoring., Data Sources: PubMed, Cochrane Library, and Scopus., Review Methods: A literature search was conducted about the normative data for PRE and HRE on multichannel intraluminal impedance-pH monitoring (MII-pH), hypopharyngeal-esophageal MII-pH (HEMII-pH), or oropharyngeal pH monitoring using PICOTS (population, intervention, comparison, outcome, timing, and setting) and PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statements. Outcomes reviewed included device characteristics, impedance/pH sensor placements, study duration, number/average and percentiles of PRE or HRE occurrence, and the event characteristics (pH, composition, and position)., Results: Of 154 identified studies, 18 met criteria for analysis, including 720 healthy individuals. HEMII-pH, MII-pH, and oropharyngeal pH monitoring were used in 7, 6, and 5 studies, respectively. The definition and inclusion/exclusion criteria of healthy individuals varied substantially across studies, with 6 studies considering only digestive symptoms to exclude potential LPR patients. Substantial heterogeneity across studies was noted, including impedance/pH sensor placements/configurations and definitions of composition (liquid, gas, mixed) and type (acid, weakly acid, nonacid) of PRE/HRE. The 95th percentile thresholds were 10 to 73 events for PRE, 0 to 10 events for HRE on HEMII-pH, and 40 to 128 for events with pH <6.0 on oropharyngeal pH monitoring. Most HREs were nonacid and occurred upright. The mean HRE among healthy individuals was 1., Conclusion: The low number of studies and the heterogeneity in inclusion criteria, definitions, and characterization of PRE and HRE limit the establishment of consensual normative criteria for LPR on ambulatory reflux monitoring. Future large multicenter studies are needed.