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…
Generative AI can serve as a draft assistant for chest radiograph reporting, accelerating workflow and promoting consistency, but requires human-in-the-loop review to mitigate hallucinations.
A DBT foundation model (DBT-DINO) improved breast density classification accuracy (79% vs 73%, p<.001) but did not significantly outperform ImageNet pretraining for 5-year breast cancer risk prediction (AUC 0.78 vs 0.76) or lesion detection (sensitivity 62% vs 67%).
A Vascular Perspective on Osteoarthritis Progression.
In advanced HCC after atezolizumab-bevacizumab failure, HAIC prolonged PFS vs TKIs (weighted median 7.1 vs 3.4 months, P<.001) and had higher response rate (35% vs 5%), with similar overall survival.
What We Left in the Chart: Rebuilding Clinical Context Using Large Language Models.
MRI Characteristics of Ovarian Serous Surface Papillary Borderline Tumor.
Foundation Models Meet Digital Breast Tomosynthesis.
Photon-counting CT: An Emerging Tool for Liver Fibrosis Assessment: An Early Career Perspective.
Hepatic Arterial Infusion Chemotherapy as Second-line Therapy in Advanced Hepatocellular Carcinoma: An Early Career Perspective.
Erratum for: Pictorial Review of Pleural Disease: Multimodality Imaging and Differential Diagnosis.
AI-enabled chamber volumetry from coronary calcium CT boosted 10-year heart failure prediction: combined model AUC 0.80 vs PREVENT 0.76 (Δ0.04, P<.001) and vs CAC score 0.70 (Δ0.10, P<.001) in 5892 asymptomatic patients. Diastolic volumes conferred higher risk than systolic volu…
Higher radiomic texture score of temporalis muscle on brain MRI was linked to increased dementia risk (HR per standard deviation: 1.32 in ADNI, 1.64 in ARIC).
A machine-learning method for dual-energy CT (DECT) elemental decomposition reduced water-equivalent range deviations by 0.4–1.4 mm and raised biological-dose gamma passing rates by up to 24.1% compared with parameterization approaches under clinically realistic noise.
A multi-stage deep learning framework (HD-TMAR) suppressed combined truncation and metal artifacts in half-detector CBCT, achieving the highest fidelity and preserving dental morphology in simulations.
A deep learning model using whole-lung CT achieved an AUROC of 0.90 (95% CI 0.86-0.94) for preoperative prediction of spread through air spaces (STAS) in lung adenocarcinoma, outperforming clinical (0.72) and radiomics (0.65) models.
A combined clinical-radiomics model predicted post-treatment PD-L1 expression in esophageal squamous cell carcinoma with a weighted AUC of 0.8214, but improvement over a clinical-only model (AUC 0.8171) was minimal. Findings are hypothesis-generating and require external validat…
A novel MRI-guided robotic catheter design demonstrated a maximum temperature rise of only 0.6°C during simultaneous imaging and 1 A actuation, staying within FDA safety limits.
The 2.5 cm Threshold Enriches Risk but Does Not Yet Define Fibrostenosis.
A CT radiomics model identified high-grade patterns (≥20%) in stage I lung adenocarcinoma with an external test AUC of 0.84. High-risk prediction was an independent prognostic factor for recurrence-free survival (HR=3.33) and overall survival (HR=1.98).
In patients with hemophilia, in vivo T2 mapping at 3 T differentiated acute hemarthrosis from nonhemorrhagic joint effusion with 100% sensitivity and 100% specificity using a 361-msec cutoff. The median T2 relaxation time was 185 msec for hemorrhagic versus 523 msec for nonhemor…
T2 Mapping Reaches the Joint: A Solution to a Classic Bedside Dilemma.
In a 780-patient study with external validation, Node-RADS and its simplified version (recSNAP) beat ESGAR criteria for MRI-based rectal cancer nodal staging (AUC 0.83 vs 0.71, P<.001). Interobserver agreement was substantial (κ=0.76).
Contrast-enhanced CT Radiomics in Lung Adenocarcinoma: Beyond Visual Enhancement.
Large Airway Osteochondroma with a "Centipede-like" Appearance.
For chest imaging diagnosis with LLM assistance, model confidence (OR 3.82) and radiologist expertise (OR 2.06) independently predict appropriate acceptance/rejection of advice, while rationale quality increases overreliance on incorrect suggestions (OR 1.71).
Large Language Models in Uroradiology: Applications, Deployment Maturity, and the Reproducibility Paradox
In 563 patients with oesophageal cancer, AI-derived 3D CT sarcopenia index independently predicted survival: high-index males 33.3 vs 21.8 months (p<0.001), females 68.8 vs 16.8 months (p<0.001), with a significant sex interaction (p=0.026).
Explainable Radiomics-Based ML Model for Lung Cancer Subtype Classification
A preoperative model integrating 18F-FDG PET hypometabolism and MRI atrophy patterns with clinical factors predicted 1-year seizure freedom after temporal lobe surgery with an AUC of 0.63–0.70, outperforming imaging-only models.
Quantitative imaging of intermuscular adipose tissue (IMAT) using CT, MRI, ultrasound, and PET provides complementary risk information beyond muscle mass alone. Evidence strongest in oncology for predicting poorer survival and treatment toxicity. Standardization needed.