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…
Advances in AI-enabled imaging may transform incidental osteoporosis findings into formally justified clinical referrals, shifting from opportunistic to diagnostic use on CT and other modalities.
A cross-sectional review assessed PCCP adoption and documentation completeness in FDA-cleared radiology AI/ML devices, using an 8-point rubric with high inter-rater reliability (ICC 0.93).
In 1591 peripheral lung lesions, CEUS-guided biopsy improved sampling, accuracy, sensitivity vs non-contrast US, especially for 3.1–5.0 cm lesions. Malignancy risk highest for atypical epithelia.
DECT iodine concentration entropy with hybrid-IR (AUC 0.94) or DLIR-M (AUC 0.91) detects chemo response in PDAC, outperforming DLIR-H (AUC 0.85, P=.02-.04).
Hybrid AI model estimated age from panoramic radiographs: overall MAE 6.52 years, but in the 31–35 age group, MAE was 0.29 years (male) and 0.16 years (female). Performance degraded in other age ranges.
A deep learning model (STA-Risk) capturing bilateral and longitudinal asymmetries on sequential mammograms achieved C-indexes of 0.72 and 0.73 in two cohorts, outperforming other risk models (0.66-0.72).
Nearly one-third of FDA-cleared radiology AI devices involve Predetermined Change Control Plans, with 65% cleared in 2025 alone. However, a new analysis finds public summaries consistently omit performance monitoring metrics and drift-trigger thresholds needed for postmarket sur…
Operationalizing Diabetic Preparation for [18F]FDG PET/CT.
Emergency Department Pediatric Capability and Image Utilization: Inequities Remain Associated With Race, Ethnicity, and Insurance Status.
Reply to "[18F]DCFPyL Versus [18F]rh-PSMA-7.3: The Jury Remains Out".
Beyond FDG, somatostatin receptor PET (e.g., DOTATATE) offers high contrast for meningioma radiotherapy planning and patient selection for peptide receptor radionuclide therapy. Amino acid PET aids in distinguishing glioma progression from treatment-related change, improving res…
MASLD affects 40% of the global population. Professional guidelines recommend screening high-risk adults with blood tests first, then noninvasive imaging to detect clinically significant fibrosis and guide management.
Correction for: "Baseline and Early Treatment MRI Model for Predicting Complete Pathologic Response to Neoadjuvant Chemoimmunotherapy in Patients With Triple-Negative Breast Cancer".
July 2026: Predeployment Artificial Intelligence (AI) Evaluation, Endometrioma Sclerotherapy, AI Assistance for Speech Recognition Errors, and More.
A random forest model using clinical, lab, and FDG PET/CT data predicted bone marrow infiltration in lymphoma with an AUC of 0.842 and accuracy of 80.9% in a retrospective testing set.
AI-assisted fracture detection in a pediatric emergency setting reduced diagnostic revisions from 8.6% to 5.7% (RR 0.66, 95% CI 0.37–1.19), but the reduction was not statistically significant. No improvement in therapeutic changes, length of stay, or diagnostic confidence.
Transhepatic Endovascular Embolization of a Splenic Vein Pseudoaneurysm after Central Pancreatectomy.
Canadian radiology comes of age: From innovation to implementation.
A Meta-learner integrating intratumoral and peritumoral MRI radiomics with clinicopathological variables predicted lymph node metastasis with AUCs of 0.915 in training, 0.895 in internal validation, and 0.875 in external testing, outperforming all base learners.
Retraction notice to "Molecular structure and mechanism of protein MSMB, TPPP3, SPI1: Construction of novel 4 pancreatic cancer-related protein signatures model based on machine learning" [Int. J. Biol. Macromol. 307 (2025) 142075].
A competency-EPA matrix framework prioritizes fundamental ultrasound image acquisition skills in POCUS curricula, addressing psychomotor training gaps and mapping competencies across examinations to guide curriculum development.
In 70 clinical fetal MRIs, automated slice-to-volume reconstruction with brain segmentation took 8.6 minutes and was rated "good" in 70% of cases. Derived supratentorial, cerebellar, and brainstem volumes strongly correlated with traditional 2D linear measurements (r=0.88–0.94),…
Computed tomography angiography (CTA) often fails to detect distal medium vessel occlusions (DMVOs), but time-to-maximum (Tmax) perfusion maps show significantly higher sensitivity, and emerging AI tools may aid diagnosis.
In vitro, endothelial cell proliferation on metals used in flow diverters and coils differed: nitinol, stainless steel, and cobalt-chromium supported growth (p<0.001), while platinum and cobalt did not. Smooth muscle cell proliferation was limited. Caution: static conditions.
A multimodal imaging workflow combining CT, 3D rotational angiography, and live fluoroscopy co-registration was technically feasible for guiding endovascular stentrode deployment in a phantom model.
Radiology's digital transformation brings musculoskeletal strain, eye strain, and burnout. A broader safety framework addressing ergonomic, visual, and organizational hazards is needed alongside radiation protection.
SPUR-iG, a deep unrolled reconstruction, yielded whole-brain 3D MR fingerprinting in under 15 seconds (111× speedup) and achieved T1 accuracy from 30-second scans matching that from 2-minute conventional scans.
A systematic review and meta-analysis analyzed anatomy-specific performance of CTA-based AI for detecting anterior circulation occlusions, including distal occlusions. The study pooled sensitivity and specificity from research published 2019–2026.
The U.S. FDA cleared Cleerly LABS (Cleerly, Inc.) under the 510(k) pathway for radiological image processing.
FDA cleared Cleerly's coronary CT angiography AI software for plaque analysis via the 510(k) pathway.