European Radiology Abdomen, a new journal from ESR/ESGAR/ESUR, has published its first articles, led by Editor-in-Chief Valérie Vilgrain. The journal addresses a growing need for dedicated abdominal imaging research, with strong early submission volume.
Mobile LDCT screening can reach vulnerable populations in resource-limited settings.
The ascending aorta and pulmonary trunk form the anterior border of the transverse pericardial sinus. This passage lies posterior to these vessels and anterior to the superior vena cava and atrial structures. It is used surgically to pass a clamp for temporary vascular control.
Study Reveals Neighborhood Socioeconomic Status Influences Brain Health
MRI Surveillance Alone Benefits Cognitive Function in Small-Cell Lung Cancer Patients
FDA Clears Innovative MR-guided PET Software for Neurological Imaging
FDA Approves Telix’s Pixclara for Glioma PET Imaging
ROOT for Reform: The Economics Behind the ROOT Act
A narrower gamma camera photopeak energy window rejects more lower-energy Compton-scattered photons; this improves contrast. Direct photons keep roughly their original energy, while scattered photons lose energy before reaching the detector.
California Western School of Law
In 3,856 CTAs, an AI tool found 39% more intracranial aneurysms missed by radiologists (55 AI-only true positives), with AI sensitivity 0.846 vs radiologist 0.718. Specificity similar (0.985 vs 0.987). Gain-to-pain ratio 1.20, best in inpatients.
Of three AI tools for intracranial hemorrhage detection on non-contrast head CT, Aidoc achieved 90.3% sensitivity and 99.0% specificity. Simulated AI+radiologist reading boosted sensitivity to 96.0% (specificity 99.4%, P<0.001), matching two-radiologist performance.
In a two-center study of 492 patients, a stacking ensemble combining CT radiomics (TabNet) and clinical features (XGBoost) predicted acute severe cholecystitis with external test AUC 0.827, outperforming standalone models (0.753–0.792). SHAP showed key predictors: neutrophil %,…
[Multimodal artificial intelligence driving precision diagnosis and treatment of otolaryngologic diseases: key challenges and future directions].
[Advances in medical image-based artificial intelligence for precision diagnosis and treatment of head and neck tumors].
Who makes the cut: Should AI-measured breast density alone stratify screening?
In transfusion-dependent thalassemia, cardiac MRI RV-pulmonary arterial coupling was higher than in controls (0.68 vs 0.59; p<0.0001) and independently associated with heart failure (OR 3.99; p=0.049).
Crisis-ready radiology: an essential component of modern healthcare security.
Radiomics in HCC: the right technology, asking the wrong questions.
AI agents for oncology—feedback-driven systems that maintain state, select tools, and revise plans under safety constraints—should target transparent and traceable clinical decision support, not autonomous cancer diagnosis, a new Viewpoint argues.
Tri-view attention deep learning classified benign versus malignant pulmonary nodules on 869 LIDC-IDRI cases with 93.08% accuracy, AUC 97.06%, and F1 92.68%, outperforming baseline models.
In 811 brain tumor patients vs 744 controls, choroid plexus volume was larger for infratentorial than supratentorial tumors (P<0.001), not independently tied to grade. Pediatric high-grade gliomas and glioneuronal/neuronal tumors exceeded adult diffuse gliomas (P<0.01).
Corrigendum to "Stacked regressions and structured variance partitioning for interpretable brain maps" [Neuroimage 298 (2024): 120772].
A multi-center study found that the NHPVS model for pulmonary vessel segmentation on chest CT achieved a Dice score of 89.2%, improving vessel length by 31.1% and branch count by 60.5% vs nnUNet-v2, with a nodule misclassification rate of 1.8%.
A new deep learning network (SSCA-DIPNet) consistently outperformed baseline methods for MRI reconstruction at up to 20× acceleration, using limited training data from two clinical datasets and the fastMRI dataset. It preserves anatomical detail and reduces artifacts.
Quantitative CCTA Plaque Differences Between Japanese and Non-Japanese Patients: Insights From ADVANCE.
Automated segmentation model GAM-DeepLabV3+ achieved a mean Dice similarity coefficient of 0.939 for renal tumors on CT, outperforming FPN and nnU-Net (P<.001), with a 95% Hausdorff distance of 1.485 pixels. External validation on KiTS19 yielded Dice 0.928.
Adding ACR TI-RADS descriptors to ultrasound images increased ChatGPT-5.4’s accuracy for thyroid nodule classification to 84.2% vs 73.8% with images alone (P=0.001), but matched features-only input (84.2%).
Osteoporosis screening: Are we entering a new era? The point of view of the clinician.
