Large Language Models (LLMs) have demonstrated strong performance on English-language medical exams, but their effectiveness in non-English, high-stakes environments is less understood. This study benchmarks nine LLMs against human examinees on the Chilean Anesthesiology Certification Exam (CONACEM), a Spanish-language board examination. A curated set of 63 multiple-choice questions was used, categorized by Bloom’s taxonomy into four cognitive levels. Model responses were assessed using Item Response Theory and Classical Test Theory, complemented by additional error analysis, categorizing errors as reasoning-based, knowledge-based, or comprehension-related. Closed-source models surpassed open-source models, with GPT-o1 achieving the highest accuracy (88.7%). Deepseek-R1 is a strong performer among open-source options. Item difficulty significantly predicted the model accuracy, while discrimination did not. Most errors occurred in application and understanding tasks and were linked to flawed reasoning or knowledge misapplication. These results underscore LLMs’ potential for factual recall in Spanish medical exams but also their limitations in complex reasoning. Incorporating cognitive classification and error taxonomy provides deeper insights into model behavior and supports their cautious use as educational aids in clinical settings.

Background/Objectives: Mandibular asymmetry is a common morphological alteration in orthodontics and orthognathic surgery, generally diagnosed with panoramic radiographs despite their limitations. Automated processing systems offer a promising alternative for improving its detection and analysis. The aim of this study was to develop a pilot computational model to detect and measure mandibular asymmetry in the body and ramus by analyzing anatomical distances in digital panoramic radiographs of adults. Methods: This was a descriptive observational pilot study, carried out on 30 digital panoramic radiographs of young adult patients (15 men, 15 women). Three craniometric points (Condylion, Gonion and Gnathion) were used as references landmarks. An algorithm was implemented in Python^® (v3.12) with OpenCV to extract anatomical coordinates and calculate Euclidean distances (Go-Gn, Co-Go) from pixels to millimeters. Data were statistically analyzed in SPSS (v23.0) using normality tests, paired t-tests, Wilcoxon tests, and Mann–Whitney U tests (p < 0.05). Results: No significant differences were observed in mandibular lengths by sex, with men having greater lengths in both the body (80.63 mm vs. 73.86 mm) and the ramus (55.82 mm vs. 49.15 mm). In addition, significant differences were found in total mandibular ramus measurements (p = 0.023). A classification of asymmetry by severity was proposed (mild: ≤3 mm, moderate: 3–6 mm, severe: >6 mm), with mild asymmetries being the most frequently found. The model showed reliable processing capacity. Conclusions: This pilot study shows the feasibility of using Python for automated measurement of mandibular asymmetry in panoramic radiographs and highlights its future potential for neural network integration and diagnostic-epidemiological use.