Radiology. Cardiothoracic imaging最新文献

Purpose To evaluate the impact of a fully automated, multitask deep learning (DL) algorithm on interreader agreement of coronary artery disease (CAD) detection and stenosis classification using coronary CT angiography (CCTA). Materials and Methods This retrospective study included CCTA examinations (n = 623 patients) performed for clinical indications on CT systems from multiple vendors between January 2010 and December 2019. An expert reader (reader 1) analyzed all CCTA scans manually and with artificial intelligence (AI)-assisted reading at the lesion, coronary segment, and patient levels using the CAD Reporting and Data System (CAD-RADS). The AI algorithm detected, quantified, and classified coronary lesions. Interreader agreement was evaluated using a second expert reader (reader 2), who analyzed a randomly selected subset of 274 patients. CAD-RADS scores from radiologist reports (reader 3) were available for 362 patients. In a subgroup of 30 patients with disagreements, R2 also interpreted the cases using AI assistance. Agreement between readings, with and without AI, was assessed using Spearman correlation, and logistic regression and mixed models evaluated the impact of AI-assisted reading on CAD-RADS classification. Results The final study sample included 11 214 coronary segments analyzed from 623 patients (mean age ± SD, 54.8 years ± 15.7; 341 male). Of these patients, 295 (47.9%) had no CAD (CAD-RADS 0), 213 (33.6%) had low risk of coronary obstruction (CAD-RADS < 3), and 115 (18.5%) had high risk of obstructive disease (CAD-RADS ≥ 3). With AI assistance, reader 1 demonstrated improved agreement with reader 2 (ρ = 0.899-0.949; P < .001) and reader 3 (ρ = 0.889-0.938; P < .001). In the subgroup with reader 1-AI disagreement, agreement between reader 1 and reader 2 was low with manual readings (ρ = 0.688) but increased substantially when both readers used AI-assisted reading (ρ = 0.975; P < .001). Conclusion AI-assisted reading using a DL algorithm significantly improved interreader agreement for CAD-RADS classification at CCTA. Keywords: Applications - CT, CT-Coronary Angiography, Deep Learning Supplemental material is available for this article. © RSNA, 2025.

Accurate cancer staging is essential for guiding treatment decisions and predicting outcomes. The TNM classification is based on three principal elements: size and extent of the primary tumor (T), the degree of spread to regional lymph nodes (N), and the presence of distant metastases (M). In thymic epithelial tumors, clinical TNM staging relies predominantly on cross-sectional imaging, particularly CT and MRI, placing radiologists at the forefront of staging accuracy. Their assessments substantially impact both clinical decision-making and the quality of data in staging registries. A major update in this field is the ninth edition of the TNM classification for thymic epithelial tumors, effective January 2025, which refines TNM category definitions to improve staging accuracy and clinical applicability. This review article outlines these updates, illustrating their application through case examples, and emphasizes the radiologist's crucial role in cancer staging, including selection of appropriate imaging techniques, interpretation of key radiologic features, and effective communication of findings to multidisciplinary teams. These insights aim to enhance staging precision and improve patient outcomes in thymic malignancies. Keywords: Thymus, Staging © RSNA, 2025.

Coronary artery calcium (CAC) is a specific marker of subclinical coronary atherosclerosis and is strongly associated with short- and long-term atherosclerotic cardiovascular disease (ASCVD) risk. Although noncontrast electrocardiographically gated cardiac CT is the reference standard for quantification of CAC (approximately 1 mSv), studies have shown that CAC can also be qualitatively interpreted and quantified on noncardiac chest CT scans with similar prognostic value. While use of dedicated CAC scans is increasing, measurement of incidental CAC represents a major untapped opportunity for ASCVD prevention, given that nearly 20 times more chest CT examinations are performed annually in the United States than dedicated CAC scans. Incidental measurement of CAC at chest CT incurs no additional cost or radiation for patients and can identify those with significant CAC burden who may be inadequately treated with ASCVD risk reduction therapies. This review outlines the fundamentals of CAC scoring, with a focus on detection and quantification of incidental CAC. It details the technical approaches and challenges of incidental CAC assessment and provides recommendations for routine reporting, clinical advisories, and subsequent patient management. The review also presents first-hand experiences from a large academic medical center's initiative to standardize incidental CAC reporting. Future directions include the use of artificial intelligence to automate both basic and advanced CAC interpretation.

Purpose To evaluate the relationship between artificial intelligence (AI)-quantified coronary plaque characteristics derived from coronary CT angiography (CCTA), stenosis severity, and high-sensitivity cardiac troponin T (hs-cTnT) levels in predicting adverse cardiovascular outcomes in emergency department patients. Materials and Methods This single-center retrospective cohort study included patients who presented acutely to the emergency department and underwent hs-cTnT testing (February 2016-March 2021). Based on peak hs-cTnT levels, patients were categorized into three groups: undetectable (<5 ng/L), intermediate (5-13 ng/L), and elevated (≥14 ng/L). All patients underwent CCTA, and those with Coronary Artery Disease Reporting and Data System score > 0 underwent plaque quantification using an AI-based plaque tool. Patients were followed up for major adverse cardiovascular events (MACE), including acute coronary syndrome, stroke, all-cause mortality, and late revascularization. Statistical analysis included nonparametric tests, χ² tests, and Cox hazards regression. Results Among 527 patients (291 [55%] male; mean age, 56 years ± 12 [SD]), 141 had undetectable, 275 had intermediate, and 111 had elevated hs-cTnT levels. Coronary artery disease prevalence at CCTA was 59% overall and 55% in patients with nonelevated hs-cTnT levels. Total, calcified, noncalcified, and low-density noncalcified plaque volumes increased significantly with higher troponin levels (P < .001). Over a median 29-month follow-up period, 22 MACE occurred. Elevated hs-cTnT level was not associated with increased MACE risk, whereas total plaque volume > 250 mm³ was a significant predictor of both MACE (hazard ratio [HR], 2.62 [95% CI: 1.13, 6.07]; P = .02) and all-cause mortality (HR, 3.62 [95% CI: 1.25, 10.50]; P = .02). Conclusion In this cohort, AI-quantified total plaque volume predicted MACE whereas troponin level did not. This study supports the use of CCTA with AI-based plaque quantification for risk stratification in a real-world population. Keywords: CT Angiography, Coronary Arteries, Arteriosclerosis, Coronary Artery Disease, Plaque Quantification, Troponin, Coronary Computed Tomography Angiography, Artificial Intelligence Supplemental material is available for this article. © RSNA, 2025.

Purpose To evaluate the clinical utility of artificial intelligence (AI) scores in years derived from chest radiographs (CXR-Lung-Risk scores) in predicting respiratory mortality in a cohort of patients with chronic obstructive pulmonary disease (COPD). Materials and Methods This retrospective single-center study included patients with COPD from a tertiary center between January 2011 and December 2015. CXR-Lung-Risk scores were derived from chest radiographs using an open-source AI algorithm. The primary outcome, respiratory mortality, was assessed for its association with CXR-Lung-Risk using a multivariable Fine-Gray model adjusted for age, sex, body mass index, smoking status, comorbidities, and pulmonary function test results. Discrimination was evaluated and benchmarked against the Global Initiative for Chronic Obstructive Lung Disease grade using the time-dependent area under the receiver operating characteristic curve (AUC). Associations between CXR-Lung-Risk and lung-function measures were examined. Results A total of 4226 patients (median age, 70 years [IQR, 63-76 years]; 3293 male) with COPD were evaluated. Respiratory mortality was observed in 19.7% (831 of 4226) of patients at a median follow-up of 6.7 years (IQR, 4.0-7.9 years). CXR-Lung-Risk was a prognostic factor for respiratory mortality (subdistribution hazard ratio per 5-year increase, 1.16; 95% CI: 1.10, 1.28; P < .001) after adjusting for lung function and clinical risk factors. Likelihood-ratio testing further confirmed its added value in multivariable models (P < .001). The AUC for CXR-Lung-Risk in predicting respiratory mortality up to 10 years was 0.76 (95% CI: 0.72, 0.79), which outperformed the Global Initiative for Chronic Obstructive Lung Disease grades (0.61; 95% CI: 0.58, 0.65; P < .001). Pulmonary function decreased with increasing CXR-Lung-Risk scores (P < .001). Conclusion This study demonstrates that CXR-Lung-Risk is a valuable open-source AI tool for predicting respiratory mortality among patients with COPD. Keywords: Chronic Obstructive Pulmonary Disease, Artificial Intelligence, Chest Radiograph, Prognostication Supplemental material is available for this article. © RSNA, 2025.

Left ventricular assist devices (LVADs) are used for short-term support, as a bridge to transplant, or as destination therapy in patients with end-stage systolic heart failure. Imaging plays a crucial role in assessing the anatomic suitability for implantation and in detecting complications following both implantation and explantation. LVAD-associated complications can affect the pump, inflow cannula, outflow graft, or driveline. Echocardiography is effective for evaluating inflow cannula position and certain parameters, such as inflow and outflow velocities, valvular regurgitation, and ventricular dilatation; however, its ability to visualize the interiors of the inflow and outflow cannulas is limited. MRI is contraindicated for patients with LVADs. Contrast-enhanced chest CT imaging has become the preferred diagnostic modality for evaluating outflow graft complications. This imaging essay describes the CT findings and complications associated with LVADs, particularly the commercially available HeartMate II and HeartMate 3 devices (Abbott Laboratories). The HeartWare device (Medtronic), although recalled by the U.S. Food and Drug Administration, will also be mentioned. Keywords: Cardiac Assist Devices, CT Imaging Supplemental material is available for this article. © RSNA, 2025.

In recent years, the landscape for the diagnosis and management of patients with aortic stenosis (AS) has rapidly changed, with a dramatic increase in therapeutic options and substantial advances in different imaging modalities. Multidetector CT (MDCT) has become an essential imaging tool for evaluating the feasibility of both surgical and interventional treatments for patients with severe AS. Novel MDCT imaging acquisition protocols, postprocessing tools, and technological advances offer not only detailed anatomic information for adequate procedural planning but also comprehensive quantitative evaluation of the myocardium for assessment of remodeling and function, both of which have prognostic and therapeutic implications. This review provides a comprehensive update on the role of novel MDCT quantitative techniques in the assessment of patients with severe AS. Keywords: CT, Cardiac, Valves, Aortic Stenosis © RSNA, 2025.

Hyperpolarized xenon 129 (¹²⁹Xe) MRI uses inhaled ¹²⁹Xe gas to visualize pulmonary function and microstructure. This review aims to summarize established and emerging quantitative measurements derived from ¹²⁹Xe MRI and MR spectroscopy (MRS) and illustrate their clinical applications in the characterization and management of cardiopulmonary diseases. They are well tolerated by adults and children with pulmonary disease, employ no ionizing radiation, and their measurements have been validated by correlation with pulmonary function tests in various cardiopulmonary diseases. ¹²⁹Xe fills unobstructed airspaces, producing three-dimensional maps of ventilation and enabling quantification of ventilation defects, dynamics, and heterogeneity. Leveraging ¹²⁹Xe's biologic solubility, gas exchange imaging and spectroscopy allow for quantification of gas transfer between airspaces, alveolar membrane, and red blood cells and are sensitive to blood oxygenation and vascular remodeling. Diffusion-weighted imaging quantifies airspace enlargement, providing models of alveolar microstructure. ¹²⁹Xe MRI can help detect early-stage disease, adding value where reference-standard tools, such as pulmonary function tests, lack sensitivity. The ability of ¹²⁹Xe MRI to assess function regionally creates opportunities for the detection of localized functional deficits and the improvement of image-guided interventions. Applications of ¹²⁹Xe MRI and MRS include planning treatment, monitoring disease progression and treatment response, and developing surrogate endpoints for clinical and therapeutic studies. Keywords: MR Imaging, MR Spectroscopy, Thorax, Lung, Hyperpolarized ¹²⁹Xe, MRI, MRS, Lung Function, Ventilation, Gas Exchange, Alveolar Microstructure © RSNA, 2025.