This case illustrates a large right ventricular myxoma causing severe right ventricular outflow tract obstruction. Multimodality imaging, including echocardiography and CT, guided the diagnosis. The patient underwent successful surgical resection, with histopathology confirming the benign tumor, leading to hemodynamic relief.

Cardiac magnetic resonance (CMR) is a cornerstone for diagnosing cardiovascular disease. However, it remains underutilized due to complex, time-consuming interpretation across multi-sequences, phases, quantitative measures that heavily reliant on specialized expertise. Here, we present BAAI Cardiac Agent, a multimodal intelligent system designed for end-to-end CMR interpretation. The agent integrates specialized cardiac expert models to perform automated segmentation of cardiac structures, functional quantification, tissue characterization and disease diagnosis, and generates structured clinical reports within a unified workflow. Evaluated on CMR datasets from two hospitals (2413 patients) spanning 7-types of major cardiovascular diseases, the agent achieved an area under the receiver-operating-characteristic curve exceeding 0.93 internally and 0.81 externally. In the task of estimating left ventricular function indices, the results generated by this system for core parameters such as ejection fraction, stroke volume, and left ventricular mass are highly consistent with clinical reports, with Pearson correlation coefficients all exceeding 0.90. The agent outperformed state-of-the-art models in segmentation and diagnostic tasks, and generated clinical reports showing high concordance with expert radiologists (six readers across three experience levels). By dynamically orchestrating expert models for coordinated multimodal analysis, this agent framework enables accurate, efficient CMR interpretation and highlights its potentials for complex clinical imaging workflows. Code is available at https://github.com/plantain-herb/Cardiac-Agent.

Background:Pulmonary artery sarcoma (PAS) is a rare malignant tumour of pulmonary vessels. Due to the lack of specific clinical symptoms, PAS is easily misdiagnosed as pulmonary thromboembolism, so that many patients miss the opportunity of early diagnosis and treatment. Case summary:A 42-year-old man presented to the hospital with a heart murmur on physical examination. After admission, he underwent serial imaging studies and was diagnosed with a pulmonary artery mass. Guided by multimodal imaging, the patient underwent surgical treatment, and post-operative pathology confirmed the diagnosis of PAS. Eighteen months after the operation, the patient's follow-up examination results indicated tumour recurrence. Discussion:This case emphasizes the advantages of combining multiple imaging modalities in the early identification of PAS and in guiding surgical management.

Aim: Constrictive pericarditis (CP) and restrictive cardiomyopathy (RCM) require distinct treatments but share overlapping clinical and imaging features, making them difficult to distinguish using single conventional imaging. This study evaluated whether machine learning models integrating multimodal imaging data could improve the differential diagnosis. Methods: CP and RCM patients were included between January 2014 and September 2024 at two hospitals. Demographic, laboratory, echocardiography, computed tomography, magnetic resonance imaging, and cardiac catheterization data were collected. Seven machine learning models were trained and validated using five-fold cross-validation. Model performance was assessed by the area under the receiver-operating characteristic curve (AUC), accuracy, sensitivity, specificity and 95% confidence interval (CI). Results: A total of 156 CP and 91 RCM patients were analyzed. Compared with RCM, CP patients showed a higher ejection fraction (62.9 ± 7.5% vs. 51.9 ± 15.2%, P < 0.001) and more frequent pericardial thickening over 4 mm (66.1% vs. 0%, P < 0.001). The Gaussian SVM of multimodal-imaging achieved the highest AUC of 0.97 (95%CI: 0.93–0.99), accuracy of 85%, sensitivity of 89%, and specificity of 90%, outperforming other SVM models, decision tree, logistic regression and k-nearest-neighbor models. ML based on multimodal imaging data achieved higher diagnostic performance than those based on any single-modality imaging features. Conclusions: A Gaussian SVM integrating multimodal imaging data markedly improves the differential diagnosis between CP and RCM. This model may help reduce misclassification and improve timely interventions for the two diseases.

Background:Acute type A aortic dissection (ATAAD) is a high-risk disease complicated by organ malperfusion (OMP), which impairs surgical outcomes. This study explored computed tomography angiography (CTA) characteristics of OMP in ATAAD patients and identified imaging biomarkers for postoperative adverse complications (AC). Methods:A retrospective analysis included 432 ATAAD patients who underwent total arch replacement with frozen elephant trunk procedure (2016-2020), excluding those without preoperative aortic CTA. CTA assessed dissection features, branch vessel involvement, and tear severity. Analyses used Kaplan-Meier (KM) curves, Cox regression, logistic regression, and restricted cubic spline (RCS). Results:Preoperative OMP was present in 155 (35.9%) patients. The Pre-OMP group had higher false lumen arc length (FLAL) ratios at multiple aortic levels, higher postoperative AC (38.7% vs. 24.5%, p = 0.002) and 30-day mortality (17.4% vs. 6.5%, p < 0.001) than the non-Pre-OMP group. KM analysis showed postoperative AC, rather than preoperative OMP, was the critical determinant of survival at follow-up (p < 0.001 vs. p = 0.069), but this effect does not persist in patients surviving beyond 1 year. Multivariable logistic regression identified ascending aorta level (AAL) FLAL ratio as an independent predictor of postoperative AC (OR: 1.335, 95%CI: 1.079-1.651, p = 0.008); RCS confirmed nonlinearity, with AC risk rising significantly when AAL FLAL ratio >0.849. Conclusions:CTA of ATAAD patients with OMP mainly shows more severe central aortic tear characteristics. Postoperative AC is a critical determinant of survival at follow-up. The AAL FLAL ratio (>0.849) can serve as a reliable predictive biomarker for postoperative AC, aiding preoperative risk stratification and individualized surgical management in ATAAD.

OBJECTIVES:To evaluate the feasibility of a non-contrast cardiac magnetic resonance (CMR)-based deep learning (DL) model for predicting left ventricular adverse remodeling (LVAR) in patients with acute ST-segment elevation myocardial infarction (STEMI). METHODS:A retrospective study included 252 patients with STEMI from two centers, randomized into training (n = 176) and testing (n = 76) cohorts. A two-stage DL framework was employed: (1) An architecture for coarse-to-fine myocardial localization and segmentation based on a 3D U-shaped network and (2) a classification model integrating imaging, morphological, and motion features extracted from cine CMR. The performance of different models was evaluated using the area under the curve (AUC), accuracy, sensitivity, specificity, and the F1 score. Regions influencing the decision-making process of the DL model were highlighted using guided gradient-weighted class activation mapping. RESULTS:The DL model demonstrated robust ability to predict LVAR, with an AUC of 0.865 (95% CI: 0.755-0.956), accuracy of 82.9%, sensitivity of 77.3%, specificity of 85.2% and F1 score 0.723 in the testing set. In multivariable analysis, conventional CMR parameters, including global longitudinal strain, left atrial reservoir strain, and infarct size, remained as independent predictors of LVAR. A combined model integrating DL features with conventional non-contrast CMR parameters improved the predictive performance (AUC: 0.889, 95% CI: 0.803-0.974 in the testing set), significantly outperforming both conventional non-contrast and contrast-enhanced CMR models. CONCLUSION:A non-contrast DL-CMR model effectively predicts LVAR in patients with STEMI, providing a gadolinium-free tool for risk stratification and personalized management.

Multimodal Large Language Models (MLLMs) have shown strong performance on public medical benchmarks, yet existing evaluations often remain weak proxies for clinical use, relying on isolated inputs and simplified recognition-style tasks. We introduce CardioLens, a leakage-resistant evaluation testbed for multi-sequence Cardiovascular Magnetic Resonance (CMR), constructed from private hospital archives through a rigorous report-to-QA construction and verification pipeline. CardioLens contains 473,896 slices and 13,494 verified QA pairs across 4D Cine, LGE, perfusion, and T2-weighted imaging, and evaluates three stages of CMR interpretation: image understanding, report generation, and disease diagnosis. Across 24 state-of-the-art MLLMs, CardioLens reveals a substantial clinical reality gap: models perform poorly overall, with performance degrading along the real CMR workflow. Confusion analysis further shows a category-collapse failure mode, where models default to frequent abnormal categories rather than distinguishing clinically distinct findings. To rule out MLLM-compatible input construction as the primary cause, we compare random, clinically motivated, and data-driven slice selection protocols under different slice budgets; performance changes only marginally, typically by about 1

BACKGROUND:The optimal surgical strategy for severe rheumatic mitral stenosis with more than mild mitral regurgitation (MR) remains unclear. We aimed to characterize mitral valve pathology in these patients and compare outcomes between valve repair and valve replacement. METHODS:This dual-center retrospective study analyzed 870 surgically treated severe rheumatic mitral stenosis patients with complete imaging. Inverse probability weighting balanced baseline characteristics (standardized mean difference <0.1). Quantitative assessment combined echocardiography and computed tomography angiography. The primary endpoint was all-cause mortality. RESULTS:Among the 870 patients, 408 (46.9%) had greater than mild MR. Compared to patients with pure stenosis, these patients had larger left atrial diameters (51.0 mm vs 49.0 mm; P < .001), higher mean pulmonary artery pressures (42.0 mm Hg vs 41.0 mm Hg; P = .015), shorter anterior (33.3 mm vs 33.7 mm; P = .026) and posterior (19.2 mm vs 20.0 mm; P < .001) leaflets, and a comparable rate of mitral valve repair (65.0% vs 63.0%; P = .595). Among patients with more than mild MR, early outcomes did not differ between repair (n = 265) and replacement (n = 143). At 5-year postoperative follow-up, repair was associated with significantly lower mortality (1.4% vs 7.1%; P = .009) and a comparable reoperation rate (1.4% vs 1.0%; P = 1.000). Survival analysis favored repair (log-rank P = .005), which independently predicted survival (hazard ratio, 0.347; P = .039). CONCLUSIONS:Nearly one-half of severe rheumatic mitral stenosis patients present with more than mild MR. Mitral repair demonstrates improved 5-year survival without an increased risk of reoperation, supporting its consideration as a key candidate strategy when feasible.

Polyps containing bizarre stromal cells are occasionally observed in the lower gynecologic tract, including the vagina, cervix, and endometrium, predominantly in perimenopausal or postmenopausal patients. These cases have traditionally been considered benign without subsequent recurrence or malignancy. We describe a rare instance of a rapid enlarging endometrial polyp characterized by atypical stromal cells in a 76-year-old postmenopausal woman, who presented with vaginal bleeding. Histologically, the polyp was noted for its abundance of atypical stromal cells interspersed among thick-walled vascular channels. Higher magnification revealed eosinophilic cytoplasm in the stromal cells, which exhibited both mono- and multinucleation, hyperchromasia with coarse chromatin, and an absence of conspicuous nucleoli and mitotic figures. A consensus among two of three consulting expert gynecological pathologists supported a benign endometrial polyp diagnosis; however, one pathologist raised the possibility of adenosarcoma, highlighting the diagnostic dilemma these unique lesions present. Despite a hysterectomy recommendation, the patient chose monitoring over immediate surgery. Persistent vaginal bleeding led to her return six months later, whereupon a 7-cm polypoid lesion in the endometrial cavity was found and removed via hysterectomy. The histopathology mirrored the initial findings, showing no myometrial invasion, prompting a re-evaluation of the presumed benign nature of the polyp given its rapid growth. Remarkably, RNA sequencing analysis of the polyp detected a JAK2::NFIB gene fusion, a novel finding for endometrial polyps with atypical stromal cells, the clinical implications of which remain to be elucidated. The rapid recurrence of the polyp within six months raises new questions about the true biological nature of these entities and the relevance of gene fusions like JAK2::NFIB in their pathogenesis, meriting further investigation.

Objective:Hyperhomocysteinemia is a risk factor for cardiovascular disease, but its impact on valve disease is lacking in research. This study was aimed at investigating the impact of hyperhomocysteinemia on rheumatic mitral valve calcification and prognosis in patients undergoing surgery. Methods:This study included 672 patients with severe rheumatic mitral valve stenosis who underwent surgery between January 2016 and December 2022. Patients were stratified by preoperative homocysteine levels. Mitral valve pathology was assessed by echocardiography and coronary CTA, with all-cause mortality as the primary mid-term endpoint. Results:Among this surgical cohort of 672 patients with severe rheumatic mitral stenosis, 208 (31.0%) patients were identified with hyperhomocysteinemia. Imaging assessment revealed that these patients had a higher Agatston score for mitral valve calcification (37.47 vs. 17.19, p = 0.038) after adjusting baseline data. Restricted cubic splines revealed a significant dose-response relationship, with mitral valve calcification increasing progressively with higher homocysteine levels (p < 0.001). The Kaplan-Meier survival analysis showed that patients with hyperhomocysteinemia had significantly lower mid-term survival rates (log-rank p = 0.004). Through univariate and multivariate COX regression analyses, it was found that hyperhomocysteinemia was an independent risk factor affecting mid-term postoperative survival (HR, 2.257; p = 0.048). Conclusions:In patients undergoing surgery for rheumatic heart disease, hyperhomocysteinemia was associated with the formation of rheumatic mitral valve calcification and increased mid-term postoperative mortality. Trial Registration: Chinese Clinical Trial Registry identifier ChiCTR2200067151.

RATIONALE AND OBJECTIVES:Ischemic left ventricular aneurysm (LVA) is associated with left ventricular thrombus (LVT), which poses a significant risk of systemic embolism. This retrospective study aimed to identify clinical and cardiac magnetic resonance (CMR)-related risk factors for LVT in patients with ischemic LVA. MATERIALS AND METHODS:This study included consecutive hospitalized patients who underwent CMR for ischemic LVA between September 2015 and June 2024. Univariate and multivariate logistic regression analyses were conducted to identify risk factors for LVT in the overall cohort and in two subgroups: acute (AMI) and chronic myocardial infarction (CMI). RESULTS:Among 384 patients, 108 (28.1%) had LVT. Multivariate logistic regression identified high-sensitivity C reactive protein (hs-CRP) levels (Odds Ratio (OR) 1.05, 95%Confidence Interval (CI) 1.01-1.08, P =0.013), apical aneurysm(OR 4.23, 95%CI 1.15-1.08, P=0.030), infarcted zone peak circumferential strain (PCS) (OR 1.14, 95%CI 1.09-1.21, P<0.001) and extent of late gadolinium enhancement (LGE) (OR 1.35 per 5% increase, 95%CI 1.21-1.50, P <0.001) as significant risk factors for LVT. In patients with AMI, hs-CRP levels, infarcted zone PCS, and extent of LGE were independent predictors of LVT. In patients with CMI, D-dimer, infarcted zone PCS, and extent of LGE emerged as significant predictors. CONCLUSION:Elevated hs-CRP levels, higher infarcted zone PCS, greater extent of LGE and apical aneurysm are independent risk factors for LVT in patients with ischemic LVA. Both infarcted zone PCS and extent of LGE are significant predictors of LVT in patients with either AMI or CMI complicated with LVA.

Background:Cardiac schwannoma is exceedingly rare, and few literature reports are available. We reported a case of primary cardiac schwannoma that performed preoperative multimodal imaging, aiming to highlight the significance of multimodal imaging evaluation and deepen our understanding of this tumour. Case summary:A 66-year-old man presented to our hospital, as physical examination revealed the presence of a mediastinal mass over a month. Subsequently, the patient underwent comprehensive examination. The images demonstrated the mass compressed the right coronary artery, but no invasion. Resection of the tumour by surgery was conducted, and the patient remains asymptomatic at follow-up. Discussion:This case emphasizes the role of multimodal imaging in preoperative evaluation of such rare cardiac tumours. For benign cardiac schwannoma, the preferred treatment is surgical resection.

OBJECTIVES:Rheumatic mitral disease remains a significant health issue, especially in developing regions. Although mitral valve (MV) repair improves outcomes, residual mild mitral regurgitation (MR) is common. This study aims to identify clinical and anatomical factors associated with residual mild MR after MV repair and to assess its impact on MR progression and survival. METHODS:A dual-center, retrospective cohort study was undertaken of 580 patients with rheumatic mitral disease who underwent MV repair successfully at 2 high-volume centers in China. Patients were classified as having no MR (n = 371) or residual mild MR (n = 209) on the basis of postoperative echocardiography. Cardiac computed tomography was used to assess MV anatomy and calcification. Logistic regression identified factors associated with residual mild MR. Long-term follow-up analyzed MR progression and survival, with propensity score matching to adjust for confounders. RESULTS:The mean age was 57.6 ± 7.5 years, and 73.8% were female. Patients with residual mild MR exhibited greater rates of MV calcification (56.0% vs 43.9%, P = .007) and subvalvular fusion (27.3% vs 19.7%, P = .045). Atrial fibrillation, systolic pulmonary artery pressure, calcification volume, and papillary muscle or chordae tendineae fusion were independently associated with residual mild MR. After 7 years, the residual mild MR group had lower freedom from progression to moderate/severe MR (66.2% vs 84.9%, P = .017) but no difference in survival (97.1% vs 96.7%, P = .98). CONCLUSIONS:Residual mild MR after MV repair is associated with an increased risk of MR progression but does not impact long-term survival during limited follow-up. Preoperative cardiac computed tomography provides valuable guidance by identifying patients at greater risk for residual mild MR.