Assessing the performance of ChatGPT and Bard/Gemini against radiologists for Prostate Imaging-Reporting and Data System classification based on prostate multiparametric MRI text reports

Objectives: Large language models (LLMs) have shown potential for clinical applications. This study assesses their ability to assign Prostate Imaging-Reporting and Data System (PI-RADS) categories based on clinical text reports. Methods: One hundred consecutive biopsy-na & iuml;ve patients' multiparametric prostate MRI reports were independently classified by 2 uroradiologists, ChatGPT-3.5 (GPT-3.5), ChatGPT-4o mini (GPT-4), Bard, and Gemini. Original report classifications were considered definitive. Results: Out of 100 MRIs, 52 were originally reported as PI-RADS 1-2, 9 PI-RADS 3, 19 PI-RADS 4, and 20 PI-RADS 5. Radiologists demonstrated 95% and 90% accuracy, while GPT-3.5 and Bard both achieved 67%. Accuracy of the updated versions of LLMs increased to 83% (GTP-4) and 79% (Gemini), respectively. In low suspicion studies (PI-RADS 1-2), Bard and Gemini (F1: 0.94, 0.98, respectively) outperformed GPT-3.5 and GTP-4 (F1:0.77, 0.94, respectively), whereas for high probability MRIs (PI-RADS 4-5), GPT-3.5 and GTP-4 (F1: 0.95, 0.98, respectively) outperformed Bard and Gemini (F1: 0.71, 0.87, respectively). Bard assigned a non-existent PI-RADS 6 "hallucination" for 2 patients. Inter- reader agreements (K) between the original reports and the senior radiologist, junior radiologist, GPT-3.5, GTP-4, BARD, and Gemini were 0.93, 0.84, 0.65, 0.86, 0.57, and 0.81, respectively. Conclusions: Radiologists demonstrated high accuracy in PI-RADS classification based on text reports, while GPT-3.5 and Bard exhibited poor performance. GTP-4 and Gemini demonstrated improved performance compared to their predecessors.