Radiology最新文献

Background Angiolymphatic invasion (ALI) is an important prognostic indicator in non-small cell lung cancer (NSCLC). However, few studies focus on radiologic features for predicting ALI in patients with early-stage NSCLCs 30 mm or smaller. Purpose To identify radiologic features for predicting ALI in NSCLCs 30 mm or smaller in maximum diameter. Materials and Methods This study was a secondary review of pathologic and CT findings from an integrated health care system between January 2016 and November 2023 for participants in the prospective study Initiative for Early Lung Cancer Research on Treatment, or IELCART. Preoperative diagnostic radiologic features possibly related to ALI, volume doubling time (VDT), and PET maximum standardized uptake value were evaluated. Multivariable logistic regression analysis, adjusted for sex, age, nodule size, and smoking status, was used to determine predictors of ALI. Model performance was analyzed with the area under the receiver operating characteristic curve (AUC). Results Of 778 resected NSCLCs 30 mm or smaller (median patient age, 69 years [IQR, 63-76 years]; 458 female patients), 715 (92%) were solid, 41 (5%) were part-solid, and 22 (3%) were nonsolid. ALI was documented in 271 (35%) resected NSCLCs, all in solid NSCLCs, representing 37.9% (95% CI: 34.4, 41.5) of solid NSCLCs. None of the 63 subsolid NSCLCs had ALI (0% [95% CI: 0, 5.75]). For the 715 solid NSCLCs (median patient age, 69 years [IQR, 63-76 years]; 420 female patients), multivariable logistic regression analysis showed that lollipop sign (odds ratio [OR] = 4.12 [95% CI: 2.82, 6.04]; P < .001) and spiculation (OR = 2.05 [95% CI: 1.42, 2.97]; P < .001) were independent predictors of ALI (AUC = 0.77 [95% CI: 0.73, 0.80]). Considering only the 474 patients in whom VDT could be calculated based on CT scans, VDT was also an independent predictor for ALI (OR = 0.96 [95% CI: 0.94, 0.98]; P < .001). Incorporating VDT into the model improved ALI prediction (AUC = 0.82 [95% CI: 0.77, 0.86]; P < .001). Conclusion For patients with NSCLCs 30 mm or smaller, ALI was present in 37.9% of solid NSCLCs and none of the 63 subsolid NSCLCs. Among solid NSCLCs, lollipop sign, spiculation, and VDT were independent radiologic predictors of ALI. © RSNA, 2025 Supplemental material is available for this article. See also the editorial by Czum in this issue.

Background Multimodality imaging is essential for personalized prognostic stratification in suspected coronary artery disease (CAD). Machine learning (ML) methods can help address this complexity by incorporating a broader spectrum of variables. Purpose To investigate the performance of an ML model that uses both stress cardiac MRI and coronary CT angiography (CCTA) data to predict major adverse cardiovascular events (MACE) in patients with newly diagnosed CAD. Materials and Methods This retrospective study included consecutive symptomatic patients without known CAD referred for CCTA between December 2008 and January 2020. Patients with obstructive CAD (at least one ≥50% stenosis at CCTA) underwent stress cardiac MRI for functional assessment. Eighteen clinical, two electrocardiogram, nine CCTA, and 12 cardiac MRI parameters were evaluated as inputs for the ML model, which involved automated feature selection with the least absolute shrinkage and selection operator algorithm and model building with an XGBoost algorithm. The primary outcome was MACE, defined as a composite of cardiovascular death and nonfatal myocardial infarction. External testing was performed using two independent datasets. Performance was compared between the ML model and existing scores and other approaches using the area under the receiver operating characteristic curve (AUC). Results Of 2210 patients who completed cardiac MRI, 2038 (mean age, 70 years ± 12 [SD]; 1091 [53.5%] female participants) completed follow-up (median duration, 7 years [IQR, 6-9 years]); 281 experienced MACE (13.8%). The ML model exhibited a higher AUC (0.86) for MACE prediction than the European Society of Cardiology score (0.55), QRISK3 score (0.60), Framingham Risk Score (0.50), segment involvement score (0.71), CCTA data alone (0.76), or stress cardiac MRI data alone (0.83) (P value range, <.001 to .004). The ML model also exhibited good performance in the two external validation datasets (AUC, 0.84 and 0.92). Conclusion An ML model including both CCTA and stress cardiac MRI data demonstrated better performance in predicting MACE than traditional methods and existing scores in patients with newly diagnosed CAD. © RSNA, 2025 Supplemental material is available for this article.

Background Ischemic late gadolinium enhancement (LGE) assessed with cardiac MRI is a well-established prognosticator in ischemic cardiomyopathy. However, the prognostic value of additional LGE parameters, such as extent, transmurality, location, and associated midwall LGE, remains unclear. Purpose To assess the prognostic value of ischemic LGE features to predict all-cause mortality in ischemic cardiomyopathy. Materials and Methods This study is a secondary analysis of a prospective dual-center trial of participants with ischemic cardiomyopathy and left ventricular ejection fraction (LVEF) under 50% referred for viability assessment using cardiac MRI between 2008 and 2022. The LGE granularity parameters (extent of ischemic LGE, transmurality, location, and associated midwall LGE) assessed by cardiac MRI experts were compared with traditional prognosticators of adverse events in ischemic cardiomyopathy (age, sex, body mass index, diabetes, smoking, dyslipidemia, heart failure hospitalization, atrial fibrillation, renal failure, known myocardial infarction, and LVEF). The primary outcome was all-cause mortality. Predictive value was evaluated using Cox regression analysis and assessed using time-dependent receiver operating characteristic curves at 10 years. The cardiac MRI LGE score was developed using LGE granularity parameters. Results Among 6082 participants (mean age, 64.5 years ± 11.8 [SD]; 4419 men), 3591 had ischemic LGE. During a median follow-up of 9.0 years (IQR, 6.6-11.5 years), 652 participants died. The presence of ischemic LGE was strongly associated with mortality (hazard ratio, 3.45 [99.5% CI: 2.55, 4.67]; P < .001). In the group with ischemic LGE, the LGE granularity model combining these LGE features showed the best predictive value above traditional prognosticators and ischemic LGE extent to predict all-cause mortality (area under the receiving operating characteristic curve [AUC] at 10 years, 0.89 [99.5% CI: 0.89, 0.90] vs 0.83 [99.5% CI: 0.83, 0.84]; P < .001). The cardiac MRI LGE score performed well in participants with ischemic LGE (AUC at 10 years, 0.87 [99.5% CI: 0.85, 0.90]). Conclusion In a large cohort of participants with ischemic cardiomyopathy, an LGE granularity model had a higher prognostic value over traditional prognosticators to predict mortality. © RSNA, 2025 Supplemental material is available for this article. See also the editorial by Lima and Ebrahimihoor in this issue.

Background Studies on the association between surveillance breast MRI in women with a personal history of breast cancer (PHBC) and advanced second breast cancer are lacking. Purpose To investigate the association between postoperative surveillance breast MRI and advanced second breast cancer in women with a PHBC by using propensity score matching (PSM). Materials and Methods Women who underwent breast cancer surgery between January 2009 and December 2014 were retrospectively identified at a single tertiary center. Second breast cancer was defined as ipsilateral or contralateral breast cancer diagnosed at least 1 year after surgery, and advanced second breast cancer was defined as second breast cancer (a) grade T2 or higher or lymph node-positive or (b) T1c triple-negative or human epidermal growth factor receptor 2-positive. Women who underwent surveillance MRI and those who did not were matched using propensity scores according to 13 clinical-pathologic characteristics. Outcomes were compared using logistic regression analysis. Results Among the 3688 women (mean age, 51.1 years ± 10.5 [SD]), 2130 underwent surveillance MRI (MRI group) and 1558 did not (non-MRI group); 1062 patient pairs were matched. Advanced second breast cancer proportions for non-MRI and MRI groups were 1.7% (27 of 1558 participants) and 0.4% (eight of 2130 participants) before PSM and 1.6% (17 of 1062 participants) and 0.7% (seven of 1062) after PSM. Surveillance MRI was associated with lower odds of advanced second breast cancer before PSM (odds ratio [OR], 0.21 [95% CI: 0.10, 0.47]; P < .001) and after PSM (OR, 0.41 [95% CI: 0.17, 0.99]; P = .048). The proportion of symptomatic second breast cancers was higher in the non-MRI group before PSM (25% [16 of 65 second cancers] vs 6.4% [three of 47]; P = .01) and after PSM (21% [10 of 48] vs 3.2% [one of 31]; P = .003). Conclusion In women with a PHBC, MRI surveillance was associated with lower odds of advanced second breast cancer before and after PSM. © RSNA, 2025 Supplemental material is available for this article.

Multi-cancer early detection (MCED) tests are already being marketed as noninvasive, convenient opportunities to test for multiple cancer types with a single blood sample. The technology varies-involving detection of circulating tumor DNA, fragments of DNA, RNA, or proteins unique to each targeted cancer. The priorities and tradeoffs of reaching diagnostic resolution in the setting of possible false positives and negatives remain under active study. Given the well-established role of imaging in lesion detection and characterization for most cancers, radiologists have an essential role to play in selecting diagnostic pathways, determining the validity of test results, resolving false-positive MCED test results, and evaluating tradeoffs for clinical policy. Appropriate access to and use of imaging tests will also factor into clinical guidelines. Thus, all clinicians potentially involved with MCED tests for cancer screening will need to weigh the benefits and harms of MCED testing, including consideration of how the tests will be used alongside or in place of other screening options, how diagnostic confirmation tests should be selected, and what the implications are for policy and reimbursement decisions. Further, patients will need regular support to make informed decisions about screening using MCED tests in the context of their personal cancer risks, health-related values, and access to care.