The 2026 Chinese lung cancer guideline adds AI and opportunistic screening in low-dose CT, expands molecular testing, and updates treatment regimens for NSCLC, including perioperative, consolidation immunotherapy, and targeted therapy for EGFR, ALK, KRAS, HER2, and MET alteratio…
On 76 test cases, the AI framework achieved Cobb angle MAE 2.4°, Lenke curve type accuracy 0.895, and reduced review time from 8.2 to 1.5 min/case.
PSAD remained below 0.11 ng/mL² across all ages in 1037 healthy German men, contrasting with continuously rising PSA. 95th percentile reference: PV 51.39 mL, PSA 2.79 ng/mL, PSAD 0.071 ng/mL².
A multiphase CT angiography spot sign identified high-risk intracerebral hemorrhage: hematoma expansion occurred in 25/51 (49%) phase-1, 13/28 (46%) phase-2, and 4/9 (44%) phase-3 spots versus 12/71 (17%) with no spot sign.
ConvNeXt-based dual-stream deep learning model combining panoramic radiographs and operator-selected CBCT slices achieved macro F1-score 0.9478 in assessing mandibular third molar-inferior alveolar canal contact (250 paired cases).
In a case-control study of 96 women with prior breast cancer and a negative screening mammogram, an AI risk model predicted subsequent breast cancer within 1 year with AUC 0.824 (95% CI 0.728-0.921), sensitivity 81.3%, specificity 76.6%.
A 3D deep learning model on post-contrast T1 MRI predicted breast cancer brain metastasis receptor status (micro-averaged AUROC 0.74; ER 0.57, PR 0.79, HER2 0.64). Proof-of-concept only, no clinical utility.
Non-contrast arterial spin labeling (ASL) MRI showed diagnostic performance comparable to DSC perfusion for distinguishing recurrent high-grade glioma from treatment effects; mean AUCs were 0.946 for ASL-only, 0.997 for DSC-only, and 0.997 for combined machine-learning models.
Food insecurity linked to higher risk of death in breast cancer
<![CDATA[From Complexity to Confidence: Reducing Cognitive Load in Modern CT Imaging]]>
American Registry of Radiologic Technologists (ARRT) opened applications for its new Imaging Assistant, Magnetic Resonance credential. Assistants will handle patient safety, positioning, and contrast support—not imaging. First exam window: Oct 15–Nov 30, 2026.
Why You Should Attend Essentials of Pediatric Imaging: A Team-Based Approach in
How much surveillance imaging is needed for meningiomas?
Patient-centered Radiology & Private Practice Leadership | RSNA Unscripted with Jennifer Kemp, MD
Wearable ultrasound is making strides.
<![CDATA[Could Photon Counting CT Enhance Local Staging of Bladder Cancer?]]>
First-ever EU Screening Week runs Sept 28–Oct 4, 2026; radiology community urged to register screening-related activities by Sept 20, 2026 to showcase imaging's role in early detection.
Member spotlight on Berat Bersu Ozcan, MD.
The most likely diagnosis in this case is?
A Rare Breast Cancer Can Read as Probably Benign on a Mammogram, and Mastectomy Is Not the Treatment It Needs
In 448 patients with newly diagnosed HCC, automated single-slice L3 muscle index and attenuation independently predicted overall survival (HR 1.63 & 1.38). Volumetric analysis did not improve discrimination (C-index 0.816 vs 0.817, p=0.11). Single-slice is a practical alternativ…
A deep learning model analyzing chest radiographs in congenital heart disease predicted pulmonary-to-systemic flow ratio (Qp:Qs) with AUC 98% for Qp:Qs>1.5 (98% sensitivity, 90% specificity) and Qp:Qs<0.9 (95% sensitivity, 92% specificity). Model concordance for increased or dec…
Explainable artificial intelligence in medical imaging: how to interpret, evaluate, and use artificial intelligence explanations
Radiologists can learn to interpret and evaluate artificial intelligence predictions using explainable AI methods including saliency maps, perturbation, concept-based reasoning, and uncertainty quantification, as outlined in a practical review.
Can AI models detect abnormality in screening mammography years before confirmed diagnosis?
Commercial AI for chest X-ray did not improve diagnostic accuracy in a prospective 5-reader crossover study; for some findings, false positives increased. However, reading time dropped by 6–17 sec (p≤0.031) and most readers reported higher confidence.
Artificial Intelligence in Radiology Training: A Latin American Perspective.
A multicentre cluster-randomised cross-over trial will evaluate whether an AI-assisted fracture detection tool (RBfracture) reduces inappropriate healthcare contacts after plain-film suspected-fracture imaging. Across 4 UK Trusts, alternating monthly 'AI on' vs 'AI off' periods…
Dynamic contrast-enhanced MRI radiomics showed preliminary potential for predicting Nottingham grade in 576 breast cancers; LASSO selection gave pooled out-of-fold AUC 0.82, while chi-squared+neural net achieved 74.78% accuracy (AUC 0.77); tumor features mattered most.
Co-produced online CPD modules for interventional radiology and cardiac catheter lab staff improved interprofessional learning and specialist knowledge, but clinician-led delivery posed sustainability challenges.
NIH Funding Patterns in Diagnostic Radiology by Subspecialty, FY2012-2025.
Just 0.2% of head/neck CT and MRI reports recommended dedicated orbital imaging. When performed within 1 year (34% adherence), 93% of follow-up exams showed pathology, most commonly neoplastic.
In 20 chemotherapy-naïve breast cancer patients with cancer-related fatigue, brain FDG-PET showed hypermetabolism in bilateral thalamus (T=-2.23, P=0.05) and hypometabolism in right anterior cingulate gyrus (T=2.28, P=0.05).
Complete occlusion of a paravertebral epidural AVF confirmed at 18 months after intermittent balloon occlusion without embolic agents. Immediate flow reduction occurred after four 10-min inflations (cumulative 40 min). A bailout when superselective access is not feasible.
