Search

Objectives: Determine if a large language model (LLM, GPT-4) can label and consolidate and analyze interventional radiology (IR) microwave ablation device safety event data into meaningful summaries similar to humans. Methods: Microwave ablation safety data from January 1, 2011 to October 31, 2023 were collected and type of failure was categorized by human readers. Using GPT-4 and iterative prompt development, the data were classified. Iterative summarization of the reports was performed using GPT-4 to generate a final summary of the large text corpus. Results: Training (n = 25), validation (n = 639), and test (n = 79) data were split to reflect real-world deployment of an LLM for this task. GPT-4 demonstrated high accuracy in the multiclass classification problem of microwave ablation device data (accuracy [95% CI]: training data 96.0% [79.7, 99.9], validation 86.4% [83.5, 89.0], test 87.3% [78.0, 93.8]). The text content was distilled through GPT-4 and iterative summarization prompts. A final summary was created which reflected the clinically relevant insights from the microwave ablation data relative to human interpretation but had inaccurate event class counts. Conclusion: The LLM emulated the human analysis, suggesting feasibility of using LLMs to process large volumes of IR safety data as a tool for clinicians. It accurately labelled microwave ablation device event data by type of malfunction through few-shot learning. Content distillation was used to analyze a large text corpus (>650 reports) and generate an insightful summary which was like the human interpretation., Competing Interests: Declaration of Conflicting InterestsThe author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

During the hands-on teaching of Interventional Radiology (IR) to Residents and Fellows, certain learning tips or guiding statements are often repeated. Over the past two decades, these tips have evolved into the "Rules for Interventional Radiology." Relying on humour and the technical and foundational principles of our subspeciality, it is hoped that these Rules for IR provide helpful guidance to learners and practising Interventionalists in their daily work.

Purpose: To provide a comprehensive overview of the literature assessing the safety and efficacy of uterine artery embolization (UAE) for patients with pedunculated subserosal fibroids., Materials and Methods: MEDLINE and Embase databases were searched without language or publication type restrictions for observational studies to estimate safety (adverse events) and efficacy (devascularization, fibroid volume reduction, and uterine volume reduction) outcomes. Case reports were included to qualitatively report adverse events. Meta-analysis was performed for single proportions and mean changes with random-effects modeling., Results: Of 98 eligible articles, 11 studies were included in the final analysis. Of the adverse events detailed in these cases, 5 events were mild, 2 were moderate (torsion of pedunculated fibroid requiring laparoscopic myomectomy and persistent bleeding after embolization requiring hysterectomy), and 1 was severe (fibroid necrosis causing bowel obstruction requiring bowel resection and hysterectomy). There were no deaths reported in the literature. The pooled risk of adverse events was 1.7% (95% confidence interval [CI], 0.29%-9.2%; 4 of 181; I 2  = 0%). The pooled devascularization rate was 75.9% (95% CI, 62.4%-85.6%; 140 of 189; I 2  = 75%) at 3.91 months of follow-up. The percent volume reduction of the dominant pedunculated fibroid was 38.6% (95% CI, 33.0%-44.2%; I 2  = 0%) at 4.3 months of follow-up. The percent uterine volume reduction was 36.7% (95% CI, 30.3%-43.0%; I 2  = 47%) at 3.5 months of follow-up., Conclusions: UAE for pedunculated subserosal fibroids has a low risk of adverse events and effectively reduces fibroid and uterine size., (Copyright © 2022 SIR. Published by Elsevier Inc. All rights reserved.)

This study assessed the safety profile of high-volume (>10 mL) 3% sodium tetradecyl sulfate (STS) sclerotherapy for the treatment of renal cysts in patients with autosomal dominant polycystic kidney disease. A total of 211 sclerotherapy treatments were performed in 169 patients over a 5-year period, with a comparison of 2 patient cohorts based on the STS volumes used. The first cohort (n = 112) received a high volume (greater than 10 mL) of STS, and the second cohort (n = 57) received a low volume (less than 10 mL). The minor adverse event rate for the cohorts was 14.5% and 9.6%, respectively (P = .310), with postprocedure pain being the most common event. One major adverse event occurred, for which the patient required hospitalization for infection after low-volume STS treatment. Doses of STS higher than those currently recommended by the Food and Drug Administration for intravascular use allow large renal cysts to be treated safely in the setting of autosomal dominant polycystic kidney disease., (Copyright © 2022 SIR. Published by Elsevier Inc. All rights reserved.)