Synthesis and quality assessment of combined time-series and static medical data using a real-world time-series generative adversarial network.

This study addresses challenges related to privacy issues in utilizing medical data, particularly the protection of personal information. To overcome this obstacle, the research focuses on data synthesis using real-world time-series generative adversarial networks (RTSGAN). A total of 53,005 data were synthesized using the dataset of 15,799 patients with colorectal cancer. The results of the quantitative evaluation of the synthetic data's quality are as follows: the Hellinger distance ranged from 0 to 0.25; the train on synthetic, test on real (TSTR) and train on real, test on synthetic (TRTS) results showed an average area under the curve of 0.99 and 0.98; a propensity mean squared error was 0.223. The synthetic and real data were similar in the qualitative methods including t-SNE and histogram analyses. The application of synthetic data in predicting five-year survival in colorectal cancer patients demonstrates comparable performance to models based on real data. This study employs distance to closest records and membership inference test to assess potential privacy exposure, revealing minimal risk. This study demonstrated that it is feasible to synthesize medical data, including time-series data, using the RTSGAN, and the synthetic data can be evaluated to accurately reflect the characteristics of real data through quantitative and qualitative methods as well as by utilizing real-world artificial intelligence models.

Gastrointestinal tract perforation after radiofrequency ablation for hepatic tumor: Incidence and risk factors.

OBJECTIVE: We analyzed the incidence and mortality rate of gastrointestinal (GI) tract perforation after radiofrequency ablation (RFA) for hepatic tumors and assess its risk factors. METHODS: This retrospective cohort study included 4799 patients with malignant tumors who underwent RFA (n = 7206). Sixty-nine cases of thermal injury to the GI tract were identified via a search of the electronic medical record system using index terms and divided into two groups according to the thermal injury with (n = 8) or without (n = 61) GI tract perforation based on follow-up CT reports. The risk factors for GI tract perforation were identified via multivariable logistic regression analysis using clinical, technical, and follow-up CT findings. RESULTS: The incidence of thermal injury to the GI tract and GI tract perforation was 0.96 % (69/7206) and 0.11 % (8/7206), respectively. The type of adjacent GI tract and history of diabetes mellitus differed significantly between the two groups (p < 0.05). The index tumor being located around the small intestine was the only significant risk factor for GI tract perforation after ablation (Odds ratio, 22.69; 95 % confidence interval, 2.59-198.34; p = 0.005 [reference standard, stomach]). All perforations were not identified on CT images immediately after RFA. The median time to detection was 20 days (range, 3-41 days). Two patients (25 %, 2/8) died due to perforation-related complications. CONCLUSION: GI tract perforation after RFA for hepatic tumors is rare; however, it is associated with high mortality. Thus, careful follow-up is required after RFA if the index tumor is located around the small intestine.

2023 Survey on User Experience of Artificial Intelligence Software in Radiology by the Korean Society of Radiology.

OBJECTIVE: In Korea, radiology has been positioned towards the early adoption of artificial intelligence-based software as medical devices (AI-SaMDs); however, little is known about the current usage, implementation, and future needs of AI-SaMDs. We surveyed the current trends and expectations for AI-SaMDs among members of the Korean Society of Radiology (KSR). MATERIALS AND METHODS: An anonymous and voluntary online survey was open to all KSR members between April 17 and May 15, 2023. The survey was focused on the experiences of using AI-SaMDs, patterns of usage, levels of satisfaction, and expectations regarding the use of AI-SaMDs, including the roles of the industry, government, and KSR regarding the clinical use of AI-SaMDs. RESULTS: Among the 370 respondents (response rate: 7.7% [370/4792]; 340 board-certified radiologists; 210 from academic institutions), 60.3% (223/370) had experience using AI-SaMDs. The two most common use-case of AI-SaMDs among the respondents were lesion detection (82.1%, 183/223), lesion diagnosis/classification (55.2%, 123/223), with the target imaging modalities being plain radiography (62.3%, 139/223), CT (42.6%, 95/223), mammography (29.1%, 65/223), and MRI (28.7%, 64/223). Most users were satisfied with AI-SaMDs (67.6% [115/170, for improvement of patient management] to 85.1% [189/222, for performance]). Regarding the expansion of clinical applications, most respondents expressed a preference for AI-SaMDs to assist in detection/diagnosis (77.0%, 285/370) and to perform automated measurement/quantification (63.5%, 235/370). Most respondents indicated that future development of AI-SaMDs should focus on improving practice efficiency (81.9%, 303/370) and quality (71.4%, 264/370). Overall, 91.9% of the respondents (340/370) agreed that there is a need for education or guidelines driven by the KSR regarding the use of AI-SaMDs. CONCLUSION: The penetration rate of AI-SaMDs in clinical practice and the corresponding satisfaction levels were high among members of the KSR. Most AI-SaMDs have been used for lesion detection, diagnosis, and classification. Most respondents requested KSR-driven education or guidelines on the use of AI-SaMDs.

Value of Sonazoid-enhanced ultrasonography in characterizing indeterminate focal liver lesions on gadoxetic acid-enhanced liver MRI in patients without risk factors for hepatocellular carcinoma.

PURPOSE: To evaluate the added value of contrast-enhanced ultrasonography (CEUS) using Sonazoid in characterizing focal liver lesions (FLLs) with indeterminate findings on gadoxetic acid-enhanced liver MRI in patients without risk factors for hepatocellular carcinoma (HCC). METHODS: Patients who underwent CEUS using Sonazoid for characterizing indeterminate FLLs on gadoxetic acid-enhanced liver MRI were. The indeterminate FLLs were classified according to the degree of malignancy on a 5-point scale on MRI and combined MRI and CEUS. The final diagnosis was made either pathologically or based on more than one-year follow-up. The diagnostic performance was assessed using a receiver operating characteristic (ROC) curve analysis, and the net reclassification improvement (NRI) was calculated. RESULTS: A total of 97 patients (mean age, 49 years ± 16, 41 men, 80 benign and 17 malignant lesions) were included. When CEUS was added to MRI, the area under the ROC curve increased, but the difference was not statistically significant (0.87 [95% confidence interval {CI}, 0.77-0.98] for MRI vs 0.93 [95% CI, 0.87-0.99] for CEUS added to MRI, P = 0.296). The overall NRI was 0.473 (95% CI, 0.100-0.845; P = 0.013): 33.8% (27/80) of benign lesions and 41.2% (7/17) of malignant lesions were appropriately reclassified, whereas 10.0% (8/80) of benign lesions and 17.6% (3/17) of malignant lesions were incorrectly reclassified. CONCLUSIONS: Although performing CEUS with Sonazoid did not significantly improve the overall diagnostic performance in characterizing indeterminate FLLs on gadoxetic acid-enhanced liver MRI in patients without risk factors for HCC, it may increase radiologist's confidence in classifying FLLs.

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.