Data set-<b>Performance of Large Language Models in Recognizing Brain MRI Sequences: A Comparative Analysis of ChatGPT-4o, Claude 4 Opus, and Gemini 2.5 Pro</b>-Diagnostics.xlsx

This dataset accompanies the study titled “Performance of Large Language Models in Recognizing Brain MRI Sequences: A Comparative Analysis of ChatGPT-4o, Claude 4 Opus, and Gemini 2.5 Pro.” It includes anonymized model outputs, image-level classification results, and task-specific accuracy data derived from 130 brain MRI images representing 13 standard sequences. Each image was analyzed by three multimodal LLMs through zero-shot prompts for five classification tasks: modality, anatomical region, imaging plane, contrast-enhancement status, and MRI sequence. Data were retrospectively collected, anonymized, and reviewed by two radiologists in consensus. These files support the quantitative findings and statistical analyses reported in the manuscript.