Academic Radiology最新文献

{"title":"Advancing Neurovascular Imaging: Optimization of Reconstruction Kernel and Quantum Iterative Reconstruction for Ultra-high-resolution Photon-Counting Detector CT Angiography of the Head and Neck","authors":"Adrienn Tóth ,&nbsp;Jennifer Yongjoo Cho ,&nbsp;John Crow ,&nbsp;Evan Wilson ,&nbsp;Kimberly Kicielinski ,&nbsp;Sami Al Kasab ,&nbsp;Jennifer Joyce ,&nbsp;Maria Gisele Matheus ,&nbsp;Eric Bass ,&nbsp;Maria Vittoria Spampinato","doi":"10.1016/j.acra.2025.07.033","DOIUrl":"10.1016/j.acra.2025.07.033","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>To find the optimal reconstruction parameters for ultra-high-resolution (UHR) photon-counting detector CT (PCD-CT) angiography of the head and neck, with a special focus on imaging intracranial aneurysms.</div></div><div><h3>Material and Methods</h3><div>18 patients with intracranial aneurysms were prospectively enrolled in this single-center study. CT angiograms were acquired in UHR mode on a clinical PCD-CT scanner. Images were reconstructed with six strength levels of a dedicated neurovascular kernel (Hv48–89) and with quantum iterative reconstruction (QIR) levels 1–4. Image noise, contrast-to-noise ratio, and vessel sharpness were determined for all reconstructions. Qualitative image quality was assessed by three readers using a 5-point Likert scale, for the best-performing reconstructions. Aneurysm dome and neck sizes were independently measured by two readers to assess inter-reader reliability.</div></div><div><h3>Results</h3><div>18 patients (mean age, 64.3 years ± 15 [SD], 5 men,) were evaluated. Three kernels (Hv56, Hv64, Hv72) were identified as best performing in the quantitative analysis. The qualitative analysis demonstrated a consistent preference for QIR level 4 across all kernels in each evaluated category (P&lt;0.001). Hv72 was the most frequently preferred, although it exhibited a higher incidence of image artifacts compared to the other kernels. Inter-reader reliability was high for dome measurements—especially on UHR images—while neck measurements showed greater variability.</div></div><div><h3>Conclusion</h3><div>Hv72 kernel with QIR level 4 proved to be the optimal combination of CTA parameters among the configurations evaluated. Our results could provide reference for protocol optimization on PCD-CT for neurovascular imaging.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 10","pages":"Pages 6093-6103"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144769259","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"LLM-Based Extraction of Imaging Features from Radiology Reports: Automating Disease Activity Scoring in Crohn’s Disease","authors":"Reza Dehdab MD ,&nbsp;Fiona Mankertz MD ,&nbsp;Jan Michael Brendel MD ,&nbsp;Nour Maalouf MD ,&nbsp;Kenan Kaya MD ,&nbsp;Saif Afat MD ,&nbsp;Shadi Kolahdoozan MD, MPH, PhD ,&nbsp;Amir Reza Radmard MD","doi":"10.1016/j.acra.2025.07.041","DOIUrl":"10.1016/j.acra.2025.07.041","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>Large Language Models (LLMs) offer a promising solution for extracting structured clinical information from free-text radiology reports. The Simplified Magnetic Resonance Index of Activity (sMARIA) is a validated scoring system used to quantify Crohn’s disease (CD) activity based on Magnetic Resonance Enterography (MRE) findings. This study aims to evaluate the performance of two advanced LLMs in extracting key imaging features and computing sMARIA scores from free-text MRE reports.</div></div><div><h3>Materials and Methods</h3><div>This retrospective study included 117 anonymized free-text MRE reports from patients with confirmed CD. ChatGPT (GPT-4o) and DeepSeek (DeepSeek-R1) were prompted using a structured input designed to extract four key radiologic features relevant to sMARIA: bowel wall thickness, mural edema, perienteric fat stranding, and ulceration. LLM outputs were evaluated against radiologist annotations at both the segment and feature levels. Segment-level agreement was assessed using accuracy, mean absolute error (MAE) and Pearson correlation. Feature-level performance was evaluated using sensitivity, specificity, precision, and F1-score. Errors including confabulations were recorded descriptively<em>.</em></div></div><div><h3>Results</h3><div>ChatGPT achieved a segment-level accuracy of 98.6%, MAE of 0.17, and Pearson correlation of 0.99. DeepSeek achieved 97.3% accuracy, MAE of 0.51, and correlation of 0.96. At the feature level, ChatGPT yielded an F1-score of 98.8% (precision 97.8%, sensitivity 99.9%), while DeepSeek achieved 97.9% (precision 96.0%, sensitivity 99.8%).</div></div><div><h3>Conclusions</h3><div>LLMs demonstrate near-human accuracy in extracting structured information and computing sMARIA scores from free-text MRE reports. This enables automated assessment of CD activity without altering current reporting workflows, supporting longitudinal monitoring and large-scale research. Integration into clinical decision support systems may be feasible in the future, provided appropriate human oversight and validation are ensured.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 10","pages":"Pages 5869-5877"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144812624","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing ESCC Grading Prediction: Tumor Markers Augmented CT Radiomics and Inflammatory Immune Model","authors":"Yanyan Xiao,&nbsp;Qingxu Liu","doi":"10.1016/j.acra.2025.05.029","DOIUrl":"10.1016/j.acra.2025.05.029","url":null,"abstract":"","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 10","pages":"Page 6352"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144856965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Post-deployment Monitoring of AI Performance in Intracranial Hemorrhage Detection by ChatGPT","authors":"Eric Rohren ,&nbsp;Mohadese Ahmadzade ,&nbsp;Sofia Colella ,&nbsp;Nina Kottler ,&nbsp;Sriyesh Krishnan ,&nbsp;Jason Poff ,&nbsp;Neelesh Rastogi ,&nbsp;Walter Wiggins ,&nbsp;Joyce Yee ,&nbsp;Carlos Zuluaga ,&nbsp;Phil Ramis ,&nbsp;Mohammad Ghasemi-Rad","doi":"10.1016/j.acra.2025.07.055","DOIUrl":"10.1016/j.acra.2025.07.055","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>To evaluate the post-deployment performance of an artificial intelligence (AI) system (Aidoc) for intracranial hemorrhage (ICH) detection and assess the utility of ChatGPT-4 Turbo for automated AI monitoring.</div></div><div><h3>Materials and Methods</h3><div>This retrospective study evaluated 332,809 head CT examinations from 37 radiology practices across the United States (December 2023–May 2024). Of these, 13,569 cases were flagged as positive for ICH by the Aidoc AI system. A HIPAA (Health Insurance Portability and Accountability Act) -compliant version of ChatGPT-4 Turbo was used to extract data from radiology reports. Ground truth was established through radiologists' review of 200 randomly selected cases. Performance metrics were calculated for ChatGPT, Aidoc and radiologists.</div></div><div><h3>Results</h3><div>ChatGPT-4 Turbo demonstrated high diagnostic accuracy in identifying intracranial hemorrhage (ICH) from radiology reports, with a positive predictive value of 1 and a negative predictive value of 0.988 (AUC:0.996). Aidoc's false positive classifications were influenced by scanner manufacturer, midline shift, mass effect, artifacts, and neurologic symptoms. Multivariate analysis identified Philips scanners (OR: 6.97, p<!--> <!-->=<!--> <!-->0.003) and artifacts (OR: 3.79, p<!--> <!-->=<!--> <!-->0.029) as significant contributors to false positives, while midline shift (OR: 0.08, p<!--> <!-->=<!--> <!-->0.021) and mass effect (OR: 0.18, p<!--> <!-->=<!--> <!-->0.021) were associated with a reduced false positive rate. Aidoc-assisted radiologists achieved a sensitivity of 0.936 and a specificity of 1.</div></div><div><h3>Conclusion</h3><div>This study underscores the importance of continuous performance monitoring for AI systems in clinical practice. The integration of LLMs offers a scalable solution for evaluating AI performance, ensuring reliable deployment and enhancing diagnostic workflows.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 10","pages":"Pages 6104-6113"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144838496","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Immersive Learning Experiences: How Augmented Reality and Virtual Reality are Shaping the Future of Radiology Education","authors":"Hadi Dahhan DO ,&nbsp;Omer A. Awan MD, MPH, CIIP","doi":"10.1016/j.acra.2024.08.033","DOIUrl":"10.1016/j.acra.2024.08.033","url":null,"abstract":"","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 10","pages":"Pages 5681-5683"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142300172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of the 2024 International Association of Pancreatology Guidelines for Identifying (Pre)Malignancy Among Presumed Intraductal Papillary Mucinous Neoplasms via CT and MRI","authors":"Wenyi Deng ,&nbsp;Chunhua Yang ,&nbsp;Fuze Cong ,&nbsp;Feiyang Xie ,&nbsp;Xiuli Li ,&nbsp;Zhengyu Jin ,&nbsp;Huadan Xue","doi":"10.1016/j.acra.2025.05.058","DOIUrl":"10.1016/j.acra.2025.05.058","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div><span>To evaluate the diagnostic performance of the 2024 International Association of Pancreatology (IAP) guidelines for presumed </span>intraductal papillary mucinous neoplasms (IPMNs).</div></div><div><h3>Materials and Methods</h3><div>We retrospectively analyzed 181 presumed IPMNs with preoperative contrast-enhanced CT and 129 presumed IPMNs with preoperative contrast-enhanced MRI. All high-risk stigmata (HRS) and worrisome features (WF) in the 2024 IAP guidelines were assessed. Multivariable logistic regression analysis developed nomograms for identifying (pre)malignancy among presumed IPMNs via CT and MRI. The diagnostic performance of nomograms was validated and compared with the 2017 IAP guidelines in independent testing cohorts.</div></div><div><h3>Results</h3><div><span><span>Elevated serum carbohydrate antigen 19–9, main pancreatic duct (MPD) ≥ 10 mm, thickened enhancing cyst wall, enhanced mural nodule or solid component, and </span>lymphadenopathy were associated with (pre)malignancy via CT and MRI. MPD ≥ 5 mm and abrupt MPD caliber changes with distal atrophy were also related to (pre)malignancy via CT. Both the CT and MRI nomograms demonstrated satisfactory and improved diagnostic performance compared to HRS in the 2017 IAP guidelines (accuracy: 77.9% vs 67.7%, </span><em>p</em> = 0.039 for CT and 75.5% vs 59.2%, <em>p</em> = 0.021 for MRI) and the six-point scale based on the 2017 version (AUC: 0.808 vs 0.726, <em>p</em> = 0.039 for CT and 0.865 vs 0.631, <em>p</em> &lt; 0.001 for MRI) in the testing cohorts. The intermodality agreement between CT and MRI was moderate to excellent, except for thickened enhancing cyst wall.</div></div><div><h3>Conclusion</h3><div>The nomograms based on the 2024 IAP guidelines effectively identified (pre)malignant lesions among presumed IPMNs and demonstrated improvement over the 2017 version when evaluated via both CT and MRI in the testing cohorts.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 10","pages":"Pages 5848-5859"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144337199","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Imaging the Future: Student Reflections on an Innovative Undergraduate Radiology Course","authors":"Lulu Liu MD ,&nbsp;Ali Shokralla ,&nbsp;Erik Venalainen MD ,&nbsp;Jonathon Leipsic MD ,&nbsp;Waqas Ahmad MD ,&nbsp;Savvas Nicolaou MD","doi":"10.1016/j.acra.2025.05.070","DOIUrl":"10.1016/j.acra.2025.05.070","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div><span>Radiology plays a critical role in modern healthcare, yet it is often underrepresented in </span>undergraduate education<span>. To address this gap, the University of British Columbia (UBC) introduced RADS 301: exploring imaging in the 21st century, the first undergraduate radiology course of its kind in Canada. This study aims to examine the demographic backgrounds, motivations, and course evaluations of students enrolled in RADS 301.</span></div></div><div><h3>Methods</h3><div>An anonymous end-of-course survey was administered to students enrolled in RADS 301 during the 2024 academic term. The survey included demographic questions, multiple-choice and Likert-scale items evaluating various aspects of the course, and an open-ended feedback section. Descriptive statistics were used to analyze categorical and ordinal data, while qualitative responses were thematically reviewed.</div></div><div><h3>Results</h3><div>Out of 187 enrolled students, 178 (95%) completed the survey. Students represented a diverse range of academic disciplines, with just over half (55%) enrolled in Health and Life Science majors and the remainder from a variety of other fields, including humanities, business, and arts. Primary motivations for taking the course included general interest in medical topics (78%) and career exploration in healthcare (52%). Course evaluations demonstrated high satisfaction across all domains, with mean ratings ranging from 4.51 (SD = 0.60) to 4.70 (SD = 0.51) on a five-point scale. The overall course rating was 4.76 (SD<!--> <!-->=<!--> <!-->0.45). Qualitative responses emphasized the engaging content, diverse instructors, impact on career clarity, and improvements in health literacy.</div></div><div><h3>Conclusion</h3><div>RADS 301 is a promising model for early exposure to radiology and interdisciplinary healthcare education at the undergraduate level. The course was well-received by students from a wide array of backgrounds and appears to enhance interest, knowledge, and confidence in radiology. Findings support the expansion of similar undergraduate courses to foster informed career exploration and promote health literacy.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 10","pages":"Pages 5665-5670"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144334407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MRI-Based Radiomic Biomarkers for Non-invasive Assessment of Liver Fibrosis in MASLD: Diagnostic Performance and Molecular Mechanisms in a Rat Model","authors":"Fangyan Li ,&nbsp;Yao Liang ,&nbsp;Xing Xia ,&nbsp;Yong Wen ,&nbsp;Maowen Tang ,&nbsp;Zhaoshu Huang ,&nbsp;Na Hu ,&nbsp;Peng Luo ,&nbsp;Pinggui Lei","doi":"10.1016/j.acra.2025.05.069","DOIUrl":"10.1016/j.acra.2025.05.069","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div><span>Accurate, non-invasive assessment of liver fibrosis<span> (LF) remains a clinical challenge. This study aimed to develop a MRI-based radiomic<span> risk score (Radscore) for staging LF and to explore the biological relevance of radiomic features using </span></span></span>transcriptomic analysis.</div></div><div><h3>Materials and Methods</h3><div><span>A total of 146 male Sprague-Dawley rats were split into two cohorts at random: 87 for training and 59 for testing. T2-weighted fat-suppressed (T2FS), proton density fat fraction (PDFF), in-phase, and out-of-phase images were among the multiparametric MRI<span> sequences obtained. After radiomic features were collected, LASSO regression and redundancy analysis were used to create a 12-feature Radscore. The achieving area under the curve (AUC) was applied to assess the diagnostic performance of the Radscore for </span></span>liver fibrosis<span> detection (F0 vs. ≥F1) and staging (≤F2 vs. ≥F3). Additionally, 32 liver tissues underwent transcriptome sequencing. Radscore-associated genes were found using Pearson correlation, weighted gene co-expression network analysis (WGCNA), and differential expression analysis. Functional enrichment analysis was then performed.</span></div></div><div><h3>Results</h3><div><span><span>The Radscore demonstrated robust diagnostic performance in detecting liver fibrosis, with AUC values of 0.90 in the training cohort and 0.89 in the testing cohort (F0 vs. ≥F1). For fibrosis staging (≤F2 vs. ≥F3), the AUCs were 0.97 and 0.96, respectively. Furthermore, the Radscore was positively correlated with 10 fibrosis-associated genes (e.g., Col1a1, Col1a2, Ptprc) involved in </span>extracellular matrix remodeling and inflammatory processes. In contrast, it exhibited negative correlations with 10 genes related to </span>mitochondrial function and vascular integrity (e.g., Ndufa7, Cox5b, Kdr).</div></div><div><h3>Conclusion</h3><div>The Radscore shows promise as a non-invasive imaging biomarker for liver fibrosis in MASLD. Its correlation with transcriptomic<span> alterations indicates potential biological relevance, establishing a foundation for future studies investigating radiogenomic connections within the framework of precision medicine.</span></div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 10","pages":"Pages 5802-5813"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144512690","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Multicentre Comparative Analysis of Radiomics, Deep-learning, and Fusion Models for Predicting Postpartum Hemorrhage","authors":"Wenzhe Zhang ,&nbsp;Xin Zhao ,&nbsp;Lingsong Meng ,&nbsp;Lin Lu ,&nbsp;Jinxia Guo ,&nbsp;Meiying Cheng ,&nbsp;Hui Tian ,&nbsp;Nana Ren ,&nbsp;Jie Yin ,&nbsp;Xiaoan Zhang","doi":"10.1016/j.acra.2025.05.068","DOIUrl":"10.1016/j.acra.2025.05.068","url":null,"abstract":"<div><h3>Rationale and Objective</h3><div>This study compared the capabilities of two-dimensional (2D) and three-dimensional (3D) deep learning (DL), radiomics, and fusion models to predict postpartum hemorrhage (PPH), using sagittal T2-weighted MRI images.</div></div><div><h3>Materials and Methods</h3><div>This retrospective study successively included 581 pregnant women suspected of placenta accreta spectrum (PAS) disorders who underwent placental MRI assessment between May 2018 and June 2024 in two hospitals. Clinical information was collected, and MRI images were analyzed by two experienced radiologists. The study cohort was divided into training (hospital 1, <em>n<!--> </em>=<!--> <!-->421) and validation (hospital 2, <em>n<!--> </em>=<!--> <!-->160) sets. Radiomics features were extracted after image segmentation to develop the radiomics model, 2D and 3D DL models were developed, and two fusion strategies (early and late fusion) were used to construct the fusion models. ROC curves, AUC, sensitivity, specificity, calibration curves, and decision curve analysis were used to evaluate the models’ performance.</div></div><div><h3>Results</h3><div>The late-fusion model (DLRad_LF) yielded the highest performance, with AUCs of 0.955 (95% CI: 0.935–0.974) and 0.898 (95% CI: 0.848–0.949) in the training and validation sets, respectively. In the validation set, the AUC of the 3D DL model was significantly larger than those of the radiomics (AUC<!--> <!-->=<!--> <!-->0.676, <em>P</em>&lt;0.001) and 2D DL (AUC<!--> <!-->=<!--> <!-->0.752, <em>P</em>&lt;0.001) models. Subgroup analysis found that placenta previa and PAS did not impact the models’ performance significantly.</div></div><div><h3>Conclusion</h3><div>The DLRad_LF model could predict PPH reasonably accurately based on sagittal T2-weighted MRI images.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 10","pages":"Pages 5930-5939"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144499011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Does Coronary CTA-Based Radiomics Have Incremental Value to Anatomic and Hemodynamic Analysis in Identifying Culprit Lesions in Patients with Acute Myocardial Infarction?","authors":"Shuo Zhang ,&nbsp;Xia Li ,&nbsp;Yan Deng ,&nbsp;Kaiyue Zhi ,&nbsp;He Zhu ,&nbsp;Peng Li ,&nbsp;Jingjing Cui ,&nbsp;Pei Nie","doi":"10.1016/j.acra.2025.06.040","DOIUrl":"10.1016/j.acra.2025.06.040","url":null,"abstract":"<div><h3>Rationale and Objectives</h3><div>Radiomics<span><span> has great potential in identifying vulnerable plaques<span> and predicting plaque progression. This study aims to investigate whether coronary CT angiography (CCTA)-based radiomics provides incremental value over anatomical and </span></span>hemodynamic<span> parameters for identifying culprit lesions in acute myocardial infarction (AMI).</span></span></div></div><div><h3>Materials and Methods</h3><div><span>This multicenter study retrospectively enrolled AMI patients who underwent CCTA within 48 h of admission. Culprit lesions were adjudicated using invasive coronary angiography and electrocardiograms. CCTA-based anatomical parameters (coronary artery disease reporting and data system [CAD-RADS], high-risk plaque [HRP]), hemodynamic parameters (fractional flow reserve derived by coronary CTA [FFR</span>_CT], change in FFR_CT across the lesion [ΔFFR_CT]), and the radiomics features of plaques were analyzed. The ability to identify culprit lesions was compared among four non-radiomics models (CAD-RADS, CAD-RADS+HRP, optimal hemodynamic, and combined anatomic-hemodynamic) and two radiomics-containing models (radiomics-only and integrated anatomic-hemodynamic-radiomics).</div></div><div><h3>Results</h3><div>Among 491 patients, 491 culprit and 1869 non-culprit lesions were analyzed. In the test cohorts 1 and 2, CAD-RADS+HRP+ΔFFR_CT<span> demonstrated the best diagnostic performance compared to other models, achieving AUCs of 0.877 (95% CI: 0.847–0.906) and 0.853 (95% CI: 0.817–0.885), respectively. CAD-RADS+HRP+ΔFFR</span>_CT significantly outperformed CAD-RADS, CAD-RADS+HRP, and Radiomics (all <em>p</em>&lt;0.05). Additionally, CAD-RADS+HRP+ΔFFR_CT+Radiomics did not outperform CAD-RADS+HRP+ΔFFR_CT (<em>p</em>&gt;0.05).</div></div><div><h3>Conclusion</h3><div>The CCTA-based anatomic-hemodynamic model accurately identifies culprit lesions in patients with AMI, while radiomics provides no significant incremental benefit.</div></div>","PeriodicalId":50928,"journal":{"name":"Academic Radiology","volume":"32 10","pages":"Pages 5733-5742"},"PeriodicalIF":3.9,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144644091","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}