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.
An AI model (EAGLE) analyzing noncontrast chest CT achieved 90.0% sensitivity for esophageal cancer and 98.5% specificity in external multicenter testing (11,466 patients), and 99.94% specificity in low-dose screening. Could enable opportunistic esophageal cancer screening.
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.
MRI of pediatric MOGAD patients admitted to the ICU frequently reveals elevated intracranial pressure, with 12 of 37 showing imaging evidence; 4 had herniation or cerebral shift.
Patients later developing pancreatic cancer lost more pancreatic volume over 5 years (9.71 vs 1.23 mL; p=0.013). Adding this marker to a basic model raised AUC from 0.679 to 0.823. Validation is needed before clinical use.
BUS TI-RADS 2.0 achieved AUC 0.98 for thyroid nodule malignancy, outperforming ACR TI-RADS (0.94) and C-TIRADS (0.93), with a 12.98% false-negative rate and 2.97% biopsy recommendation rate.
AI-Enabled Care Pathway: Shifting the Paradigm in Prostate Cancer Diagnosis
A deep learning model predicted brain tumor contrast enhancement from non-contrast MRI alone: 83% accuracy, 92% sensitivity, 74% specificity on over 1,100 scans, but authors say it is not yet accurate enough to replace gadolinium-enhanced MRI.
<![CDATA[Radiology Roundup of FDA News: September 13 — September 19]]>
Small Bowel Disease: CT Findings
Select Radiology Solutions: Remote General Radiologist- Various Employment Schedules available with Lucrative Compensation! Job Opening in Remote
Lessons in Leadership: Paving the Path for Others.
Thoracic duct embolization achieved clinical success in 62.5% (25 of 40) of persistent postoperative chylothorax after complete ligation; day 1 drain output was lower in successes (324 vs 852 mL, p=0.0003).
Bilateral Transarterial Microembolization for Primary Ankle Osteoarthritis: A Systematic Multi-Vessel Approach Targeting the Perimalleolar Arterial Network.
Cara Veterinary Employer Profile - American College of Radiology Career Center
Beth Israel Lahey Health Employer Profile - American College of Radiology Career Center
ML analysis of CT-based nasal and maxillary sinus measurements achieved AUC 0.771 and 74.4% accuracy for sex prediction in 195 Egyptians. Nasofrontal angle and nasion-tip distance were top predictors.
A multimodal model integrating clinical, radiomic, and deep learning features from MRI achieved a macro-average AUC of 0.912 for classifying pediatric pineal region tumors (germinoma, NGGCT, PPT), outperforming single-modality models.
At 0.55T, liver flip angle variation was only 98.2±2.9% vs 94.6±7.4% at 1.5T and 100.7±9.8% at 3T, making B1+ correction unnecessary. Combined RF spoiling correction and deep learning reconstruction enabled accurate T1 mapping.
Streamlining Radioembolization without mapping angiography for Cholangiocarcinoma or Metastatic Liver Tumors ≤7 cm Using Yttrium-90 Resin Microspheres: Protocol for a Multicenter Prospective Registry Study (ISTAR-03).
Photon-counting CT reduced nondiagnostic coronary stents to 1.1% vs 46.8% for energy-integrating detector CT and achieved 100% sensitivity and negative predictive value for detecting ≥50% in-stent restenosis, with per-patient accuracy 90.48%.
In a retrospective two-center study, a CT habitat-based radiomics model combined with clinical factors predicted occult central node metastases in papillary thyroid cancer, achieving an AUC of 0.878 on external validation.
A deep learning MRI system distinguished six glioma/glioneuronal tumor subtypes with external test AUCs 0.69–0.92 and accuracy 0.81–0.95 in 1,844 patients, best for pilocytic astrocytoma versus pleomorphic xanthoastrocytoma (AUC 0.95).
Brain 3D-FLAIR showed near-perfect agreement (Cohen's Kappa 0.961) with dedicated optic nerve 2D-STIR in detecting optic nerve lesions in clinically isolated syndrome, with accuracy 93.1% vs 91.2%. It may serve as an alternative in routine brain MRI.
Member Spotlight: Nabawanuka
57-year-old woman with chronic shortness of breath revisited
DeepHealth gets FDA nod for Chest XRay AI tool
<![CDATA[Foundation Model-Based Chest X-Ray AI Software Gets FDA Nod]]>
Carestream Healthcare launches Starview digital radiography system
Koning Breast CT System receives Australian TGA approval
RSNA 2026 Plenaries Highlight Radiology’s Central Role in Healthcare
