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.
Noncontrast CT-based unimodal imaging AI detected anterior circulation large vessel occlusion with sensitivity 0.78 (95% credible interval 0.68-0.86) and specificity 0.88, versus expert readers' sensitivity 0.62 and similar specificity; low-certainty retrospective evidence.
SIR Foundation research consensus panel calls for standardized acute DVT intervention endpoints: biologically informed patient selection, anatomic/physiologic metrics, validated severity tools, longitudinal biomarkers, and economic/QoL measures to advance precision and reduce po…
EcoRad: Integrating Sustainability and Economics in Radiology
Bracco Imaging Secures FDA Approval and Health Canada Authorization for VueJect®, an Innovative Ultrasound Contrast System
Are respiratory physicians adherent to guidelines when evaluating pulmonary nodules?
In 46 pregnant participants, spontaneous non-labor uterine contractions reduced placental T2* by 4.64±0.75 ms, while maternal hyperoxia increased it by 11.36±0.62 ms. Contraction-induced T2* reduction and deformation intensified with gestational age and duration.
Explainable Artificial Intelligence In Oral And Maxillofacial Radiology: From Black-Box Algorithms To Clinically Trustworthy Imaging
Artificial Intelligence In Radiology Reporting: The Role Of Natural Language Processing And Large Language Models
Imaging Clinically Significant Fibrosis: A New Frontier for Radiologists in MASLD.
Caution Needed When Applying Vision-Language Models for Chest Radiography Nodule Screening.
The Role of Visual Language Models in the Management of the Axilla in Breast Cancer.
Important Reminders When Using Propensity-Score Matching.
Small and pure ground-glass lung nodules are the ones most often missed on chest computed tomography (CT), especially near vessels or airways. Reader fatigue and bias also contribute. Knowing these patterns helps radiologists use technical and cognitive strategies to reduce miss…
Why Our Teaching Rotations Will Remain in Person.
Radiology Economics Forecast: Is It Doom and Gloom?
In 175 patients, a preoperative MRI connectome model predicted long-term seizure recurrence after temporal lobe epilepsy surgery with 80.0% specificity and 63.9% negative predictive value, outperforming clinical variables.
An ECG-only deep-learning model predicted respiratory-induced liver motion with a mean absolute error of 2.83 mm during normal breathing and 4.02 mm overall, with correlation coefficients above 0.90 in eight volunteers.
AI volumetric segmentation after vestibular schwannoma radiosurgery tracked internal regrowth with median Dice 0.85 vs expert reads; internal regrowth was just 4.7% of cases but accounted for 27% of radiographic failures (170 MRI scans, 85 patients).
FAPI PET/CT showed 93.0% sensitivity vs 76.5% for FDG (p=0.002) in 176 patients with suspected HPB cancers. Nodal accuracy was 86.8% vs 71.1%. Dual-tracer approach did not improve accuracy. Prospective validation needed.
In chronic subdural hematoma, a contralateral middle meningeal artery (MMA) larger than the ipsilateral by ≥0.2 mm predicted additional collaterals: 93.75% vs 35.78% (p<0.001). Ipsilateral MMA ≤1.384 mm also predicted collaterals (p<0.001).
Similar 90-day functional outcomes for tenecteplase vs alteplase before thrombectomy in patients ≥80 with large-vessel occlusion stroke (mRS ≤3: 48.0% vs 43.5%; OR 1.20, 95% CI 0.86–1.67).
In a 77-patient cohort with knee osteoarthritis, adding Lipiodol to imipenem/cilastatin genicular artery embolization did not improve long-term outcomes; early pain reduction at 1 month was greater (p=0.024). Both regimens were safe.
Only one-third of patients received osteoporosis medication after vertebral augmentation. The incidence of new vertebral fractures fell from 0.519 to 0.076 per person-year after starting medical management (incidence rate ratio 6.79, 95% CI 3.37-15.51), a 6-fold decreased risk.
Survival According to Radiomic and Visual UIP Classification in Fibrotic ILD.
Low-molecular-weight heparin (LMWH) remains first-line for venous thromboembolism in pregnancy and breastfeeding; vitamin K antagonists are contraindicated at 6–12 weeks and third trimester, and direct oral anticoagulants are contraindicated throughout.
Autonomous ultrasound systems using AI and robotics are now capable of self-directed image acquisition, autonomous probe adjustment, and real-time anatomy identification, potentially reducing operator variability and expanding access to ultrasound in underserved regions.
Image-fusion-guided focal irreversible electroporation for prostate cancer: Step-by-step technique.
AI combining mammography and tomosynthesis radiomics+deep learning (DL) improved breast microcalcification differentiation: test AUC 0.835, significantly better than BI-RADS (0.809) or single-modality models (P<0.05).
FDA cleared BeamWorks Inc.'s CadAI-B Dx, an AI-based diagnostic software for radiology, via the 510(k) pathway.
FDA clears Fujifilm’s BA-004CC fluoroscopic x-ray system via the 510(k) premarket notification pathway.
FDA clears Cortechs Labs' NeuroQuant PET software (product code LLZ) via 510(k) pathway for quantitative analysis of brain PET images.
In 300 forensic long-bone radiographs, ChatGPT-5.2 achieved 83.0% accuracy (sensitivity 70.0%, specificity 96.0%), but all three vision language models showed unstable repeat reads one month later (within-model kappa 0.131-0.622). Reproducibility, not just accuracy, needs benchm…