Open AI Chat GPT as a Patient-Facing Translator: Evaluating Artificial Intelligence-Generated Spanish and English Summaries of Breast Imaging Reports

Objective: Breast imaging reports are typically written for healthcare professionals using technical language that limits patient comprehension.This study evaluated ChatGPT's ability to simplify breast radiology reports into Spanish while maintaining clinical accuracy. Methods:We collected 50 publicly posted breast imaging reports from Reddit, an online discussion platform, and used GPT-4o (March 2025 version) to generate simplified versions in both English and Spanish.Readability was assessed using Flesch Reading Ease Score (FRES) and Flesch Kincaid Grade Level (FKRL) for English outputs and Fernndez-Huerta and Szigriszt-Pazos indices for Spanish outputs.A bilingual, fellowship-trained academic breast radiologist reviewed all outputs for clinical accuracy.Analysis included paired t-tests and Wilcoxon signed-rank tests. Results:ChatGPT significantly improved English readability, reducing FKRL from 11.6 1.8 to 9.6 1.1 and increasing FRES from 33.9 11.8 to 59.7 5.4 (both p < .001).Word count was slightly higher for English outputs (248 80 vs. 240 140 words, p = .03),indicating gains were due to language simplification rather than content reduction.Spanish translations achieved Fernndez-Huerta and Szigriszt-Pazos scores of 87.0 5.1 and 83.5 5.2, respectively both in the "easy" readability range with no outputs exceeding the equivalent of an 8th-grade FKRL score.All English and Spanish outputs (100%) preserved clinical accuracy.Subgroup analysis showed that even the most complex reports, such as those involving biopsy, were effectively simplified without loss of meaning. Conclusion:ChatGPT enhances the readability of breast imaging reports in English and Spanish while preserving fidelity.