Using GPT-4 for LI-RADS feature extraction and categorization with multilingual free-text reports.

BACKGROUND AND AIMS: The Liver Imaging Reporting and Data System (LI-RADS) offers a standardized approach for imaging hepatocellular carcinoma. However, the diverse styles and structures of radiology reports complicate automatic data extraction. Large language models hold the potential for structured data extraction from free-text reports. Our objective was to evaluate the performance of Generative Pre-trained Transformer (GPT)-4 in extracting LI-RADS features and categories from free-text liver magnetic resonance imaging (MRI) reports. METHODS: Three radiologists generated 160 fictitious free-text liver MRI reports written in Korean and English, simulating real-world practice. Of these, 20 were used for prompt engineering, and 140 formed the internal test cohort. Seventy-two genuine reports, authored by 17 radiologists were collected and de-identified for the external test cohort. LI-RADS features were extracted using GPT-4, with a Python script calculating categories. Accuracies in each test cohort were compared. RESULTS: On the external test, the accuracy for the extraction of major LI-RADS features, which encompass size, nonrim arterial phase hyperenhancement, nonperipheral 'washout', enhancing 'capsule' and threshold growth, ranged from .92 to .99. For the rest of the LI-RADS features, the accuracy ranged from .86 to .97. For the LI-RADS category, the model showed an accuracy of .85 (95% CI: .76, .93). CONCLUSIONS: GPT-4 shows promise in extracting LI-RADS features, yet further refinement of its prompting strategy and advancements in its neural network architecture are crucial for reliable use in processing complex real-world MRI reports.

Transperineal versus transrectal prostate fiducial insertion in radiation treatment of prostate cancer: a systematic review and meta-analysis.

PURPOSE: To provide more accurate and definitive conclusions regarding the clinical and technical complications associated with the transperineal (TP) and transrectal (TR) approaches, a comprehensive review of observational studies and randomized controlled trials was conducted. This systematic review covered all eligible studies to facilitate a thorough comparison of complications linked to the two fiducial marker insertion methods, TP and TR. METHODS: A comprehensive search of the literature was conducted, encompassing databases such as PubMed, Embase, and the Cochrane Library, up to July 7, 2023. The relative risk and 95% confidence interval were utilized to evaluate the diagnosis and complication rates. RESULTS: The final selection for the methodological quality analysis included 13 observational studies that utilized TP and TR gold fiducial insertion approaches. The meta-analysis revealed significantly lower risks of urinary tract infections (UTI) and rectal bleeding with the TP approach. CONCLUSION: The use of both TP and TR techniques for placing gold seed fiducial markers has proven to be an effective, safe, and well-tolerated method for image-guided radiation therapy in prostate cancer patients. A significant benefit of the TP technique is its ability to avoid rectal puncture, thereby reducing the risk of UTIs. Although the incidence of UTIs and rectal bleeding associated with the TR method is relatively low, these complications can disrupt patient wellbeing and potentially cause delays in treatment.