Repository Collection:

Artificial Intelligence-Driven Drafting of Chest X-Ray Reports: 2025 Position Statement From the Korean Society of Thoracic Radiology Based on an Expert Survey

2026-09-18T07:32:54Z

Title: Artificial Intelligence-Driven Drafting of Chest X-Ray Reports: 2025 Position Statement From the Korean Society of Thoracic Radiology Based on an Expert Survey All Author(s): Won Gi Jeong; Eui Jin Hwang; Gong Yong Jin; Ju-Hyung Lee; Se Ri Kang; Hongseok Ko; Bomi Gil; Jin Hwan Kim; Tae Jung Kim; Chan Ho Park; Kyongmin Sarah Beck; Min Ji Son; Jeong Joo Woo; Seung-Jin Yoo; Jin Young Yoo; Soon Ho Yoon; Ji Won Lee; Kyung Nyeo Jeon; Yeon Joo Jeong; Soo-Youn Ham; Su Jin Hong; Wonju Hong; Jin Mo Goo Abstract: Objective: Generative artificial intelligence (AI) systems can be used to draft automated chest X-ray (CXR) reports. Although promising in terms of efficiency and workforce shortages, their accuracy, reliability, and clinical utility remain uncertain. This article presents the Korean Society of Thoracic Radiology (KSTR) position statement on AI-assisted CXR report drafting, derived from a Delphi survey of experts who used the software on a modest case set. Materials and methods: Twenty thoracic radiologists completed a Delphi survey after reviewing 60 CXR cases using an AI-based tool for automated report drafting (KARA-CXR, version 1.0.0.3; KakaoBrain, Seoul, Republic of Korea). Prior to the Delphi survey, the participants individually reviewed 60 CXR cases at their respective workplaces as part of the survey preparation process. The 60 cases were distributed evenly across six clinical settings (health screening, inpatient, emergency department, intensive care unit, respiratory outpatient, and non-respiratory outpatient), with 10 cases in each setting. The participants individually selected CXR cases in which they had worked. The entire selection and review processes were completed within 1 month. Subsequently, two Delphi rounds were conducted. Participants rated 12 key questions (72 items) regarding the clinical applicability of the AI-based tool on a 9-point Likert scale. Consensus required ≥70% agreement. Results: Consensus emerged for 41 of 72 items (56.9%). Respondents adopted a neutral stance on most questions concerning accuracy and clinical integration; they were neither impressed nor disappointed with the tool. A favorable view emerged only for health-screening examinations. Conversely, the stand-alone use of the AI-based tool in routine practice was opposed. Participants stressed the need for further performance optimization before deployment and advocated society-endorsed education and guidelines before adoption. Conclusion: The KSTR supports the use of an AI-based automated CXR report-drafting tool only in health-screening settings with radiologist validation and opposes its standalone use in routine practice, recommending performance optimization and society-endorsed education and guidelines before its adoption.

Diagnostic Accuracy and Clinical Value of a Domain-specific Multimodal Generative AI Model for Chest Radiograph Report Generation

2026-09-18T07:32:51Z

Title: Diagnostic Accuracy and Clinical Value of a Domain-specific Multimodal Generative AI Model for Chest Radiograph Report Generation All Author(s): Eun Kyoung Hong; Jiyeon Ham; Byungseok Roh; Jawook Gu; Beomhee Park; Sunghun Kang; Kihyun You; Jihwan Eom; Byeonguk Bae; Jae-Bock Jo; Ok Kyu Song; Woong Bae; Ro Woon Lee; Chong Hyun Suh; Chan Ho Park; Seong Jun Choi; Jai Soung Park; Jae-Hyeong Park; Hyun Jeong Jeon; Jeong-Ho Hong; Dosang Cho; Han Seok Choi; Tae Hee Kim Abstract: Background Generative artificial intelligence (AI) is anticipated to alter radiology workflows, requiring a clinical value assessment for frequent examinations like chest radiograph interpretation. Purpose To develop and evaluate the diagnostic accuracy and clinical value of a domain-specific multimodal generative AI model for providing preliminary interpretations of chest radiographs. Materials and Methods For training, consecutive radiograph-report pairs from frontal chest radiography were retrospectively collected from 42 hospitals (2005–2023). The trained domain-specific AI model generated radiology reports for the radiographs. The test set included public datasets (PadChest, Open-i, VinDr-CXR, and MIMIC-CXR-JPG) and radiographs excluded from training. The sensitivity and specificity of the model-generated reports for 13 radiographic findings, compared with radiologist annotations (reference standard), were calculated (with 95% CIs). Four radiologists evaluated the subjective quality of the reports in terms of acceptability, agreement score, quality score, and comparative ranking of reports from (a) the domain-specific AI model, (b) radiologists, and (c) a general-purpose large language model (GPT-4Vision). Acceptability was defined as whether the radiologist would endorse the report as their own without changes. Agreement scores from 1 (clinically significant discrepancy) to 5 (complete agreement) were assigned using RADPEER; quality scores were on a 5-point Likert scale from 1 (very poor) to 5 (excellent). Results A total of 8 838 719 radiograph-report pairs (training) and 2145 radiographs (testing) were included (anonymized with respect to sex and gender). Reports generated by the domain-specific AI model demonstrated high sensitivity for detecting two critical radiographic findings: 95.3% (181 of 190) for pneumothorax and 92.6% (138 of 149) for subcutaneous emphysema. Acceptance rate, evaluated by four radiologists, was 70.5% (6047 of 8680), 73.3% (6288 of 8580), and 29.6% (2536 of 8580) for model-generated, radiologist, and GPT-4Vision reports, respectively. Agreement scores were highest for the model-generated reports (median = 4 [IQR, 3–5]) and lowest for GPT-4Vision reports (median = 1 [IQR, 1–3]; P < .001). Quality scores were also highest for the model-generated reports (median = 4 [IQR, 3–5]) and lowest for the GPT-4Vision reports (median = 2 [IQR, 1–3]; P < .001). From the ranking analysis, model-generated reports were most frequently ranked the highest (60.0%; 5146 of 8580), and GPT-4Vision reports were most frequently ranked the lowest (73.6%; 6312 of 8580). Conclusion A domain-specific multimodal generative AI model demonstrated potential for high diagnostic accuracy and clinical value in providing preliminary interpretations of chest radiographs for radiologists.

Ultrasound-guided genitofemoral nerve block for femoral arterial access gain and closure: a randomized controlled trial

2026-08-13T05:35:15Z

Title: Ultrasound-guided genitofemoral nerve block for femoral arterial access gain and closure: a randomized controlled trial All Author(s): Youngjong Cho; Sung-Joon Park; Hyoung Nam Lee; Sangjoon Lee; Woong Hee Lee; Seung Soo Kim; Nam Hun Heo Abstract: Objectives: This study aimed to compare the analgesic efficacy and safety of the femoral branch block of the genitofemoral nerve (FBB) versus local infiltration anesthesia (LIA) for femoral arterial access gain and closure. Methods: Eighty-two patients (age, 64.8 ± 10.9 years; female, 30.5%) undergoing endovascular procedures using 5-Fr femoral sheath were assigned to either FBB (n = 41) or LIA (n = 41). In both groups, 2% lidocaine HCL with 1:100,000 epinephrine was used as an anesthetic solution. Pain scores during access gain and closure were evaluated using a visual analog scale (score 0-10), patient satisfaction levels with the quality of anesthesia were scored on a 7-point Likert scale, and adverse events were recorded. Results: The primary endpoint, pain scores during access closure, was significantly lower in the FBB group than in the LIA group (0.1 ± 0.37 vs 1.73 ± 0.92; p < 0.001). The FBB group also had significantly lower pain scores during access gain compared to the LIA group (0.83 ± 0.83 vs 2.78 ± 1.26; p < 0.001). There was an inverse relationship between pain scores and FBB after adjustment for age, gender, and body mass index (p < 0.001). FBB group reported significantly higher satisfaction with anesthesia quality compared to the LIA group (6.49 ± 0.64 vs 4.05 ± 1.05; p < 0.001). No complications were recognized in either group. Conclusions: Ultrasound-guided genitofemoral nerve blocks offered better acute pain relief and higher patient satisfaction than LIA during femoral arterial access gain and closure. Clinical relevance statement: In this prospective randomized controlled trial, ultrasound-guided genitofemoral nerve blocks offered better acute pain relief than local infiltration anesthesia, resulting in enhanced patient satisfaction. Key points: • FBB provided better pain relief during access gain and closure than LIA. • FBB offered higher patient satisfaction with the quality of anesthesia than LIA. • No anesthesia-related or access site complications were recognized in either treatment group.

RETRACTED: Brachial plexus block using only 1% lidocaine to reduce pain during the endovascular treatment of dysfunctional arteriovenous access

2025-12-03T07:38:56Z

Title: RETRACTED: Brachial plexus block using only 1% lidocaine to reduce pain during the endovascular treatment of dysfunctional arteriovenous access All Author(s): Sung-Joon Park; Hwan Hoon Chung; Yun Hak Lee; Hyoung Nam Lee; Youngjong Cho; Sangjoon Lee; Seung Hwa Lee; Woo Young Yang Abstract: No abstract available