Translating musculoskeletal radiology reports into patient-friendly summaries using ChatGPT-4.

OBJECTIVE: To assess the feasibility of using large language models (LLMs), specifically ChatGPT-4, to generate concise and accurate layperson summaries of musculoskeletal radiology reports. METHODS: Sixty radiology reports, comprising 20 MR shoulder, 20 MR knee, and 20 MR lumbar spine reports, were obtained via PACS. The reports were deidentified and then submitted to ChatGPT-4, with the prompt "Produce an organized and concise layperson summary of the findings of the following radiology report. Target a reading level of 8-9th grade and word count <300 words." Three (two primary and one later added for validation) independent readers evaluated the summaries for completeness and accuracy compared to the original reports. Summaries were rated on a scale of 1 to 3: 1) summaries that were incorrect or incomplete, potentially providing harmful or confusing information; 2) summaries that were mostly correct and complete, unlikely to cause confusion or harm; and 3) summaries that were entirely correct and complete. RESULTS: All 60 responses met the criteria for word count and readability. Mean ratings for accuracy were 2.58 for reader 1, 2.71 for reader 2, and 2.77 for reader 3. Mean ratings for completeness were 2.87 for reader 1 and 2.73 for reader 2 and 2.87 for reader 3. For accuracy, reader 1 identified three summaries as a 1, reader 2 identified one, and reader 3 identified none. For the two primary readers, inter-reader agreement was low for accuracy (kappa 0.33) and completeness (kappa 0.29). There were no statistically significant changes in inter-reader agreement when the third reader's ratings were included in analysis. CONCLUSION: Overall ratings for accuracy and completeness of the AI-generated layperson report summaries were high with only a small minority likely to be confusing or inaccurate. This study illustrates the potential for leveraging generative AI, such as ChatGPT-4, to automate the production of patient-friendly summaries for musculoskeletal MR imaging.

CT and MR imaging surveillance of stage 1 renal cell carcinoma after microwave ablation.

OBJECTIVE: To describe the CT and MR imaging findings after microwave ablation of clinical stage 1 renal cell carcinoma (RCC). METHODS: This single-center retrospective study was performed under a waiver of informed consent. 49 patients (38 M/11F, mean age 66 ± 9.0) with 52 cT1a RCC and 19 patients (10M/9F, mean age 67 ± 9.7) with 19 cT1b RCC were treated with percutaneous microwave ablation between January 2012 and June 2014. The size and volume of the RCC and ablation zone were measured and the kidney, ablation zones and retroperitoneum were assessed at immediate post-procedure CT and surveillance CT and MRI. RESULTS: Median imaging follow-up was 18 months (IQR 12-28). Ablation zones were heterogeneously hyperintense on T1W and hypointense on T2W MRI and hyperdense at CT. Thin peripheral, but no internal enhancement after contrast administration signified successful ablation zones. Ablation zones decreased in size, but did not resolve during surveillance. Immediate post-procedure subcapsular gas and hematoma (5/71, 7%) resolved prior to first follow-up. Focal, enhancing soft tissue within the ablation zone, invariably along the renal margin, signified local recurrence. Local recurrence rates were higher for T1b (2/19, 11%) compared to T1a (1/52, 2%). Urinomas (4/71, 6%) decreased in size and resolved during surveillance. Retroperitoneal fat necrosis (6/71, 9%), with opposed-phase loss of T1W MRI signal, was confirmed at histology after percutaneous biopsy. CONCLUSION: CT and MR imaging features after microwave ablation of renal cell carcinoma are predictable and reliably demonstrate treatment success, early and delayed complications, and local recurrences that can guide patient management.