MRI radiomics predicts pituitary adenoma consistency, molecular subtype, and recurrence often with AUCs >0.80, but clinical translation awaits multicenter validation.
AI-based image reconstruction and postprocessing can make images look sharper but may hide diagnostic failures. Understanding these limitations and failure modes is critical for safe use.
A Case of Anterior Tibial Artery Occlusion Following Transarterial Embolization for Plantar Fasciitis via Retrograde Tibiopedal Access.
AI enables opportunistic BMD assessment from radiography, CT, and MRI; CT deep learning yields correlations of 0.8-0.9 versus reference quantitative CT. Barriers: interoperability, reimbursement, model explainability.
An automated tool (Auto-Bone-Caliper) segmented all 19 metacarpal and phalangeal bones in pediatric hand radiographs with Dice scores of 98.7% (normal) and 95.0% (dysmorphic); length errors were 2.2% and 4.5%.
A deep learning model (BiLSTM + attention + transformer) achieved 99.8% accuracy for lung cancer detection from clinical text (including radiology reports) in an experimental study. Training time was 13.99 s. Small dataset and overfitting concerns limit generalizability.
DeepSeek-R1 (open-source) outperformed ChatGPT-o1 (proprietary) in academic writing quality for neuroradiology (mean Likert 3.23 vs 3.02, p=0.021). However, citation fabrication was higher (36.7% vs 2.6%). Both models require rigorous human review.
Hologic received FDA 510(k) clearance for its I-View mammography system. The device is classified under product code MUE (Full Field Digital Mammography).
Correction: Malignant cerebral edema after endovascular thrombectomy: a multimodal prediction model based on post-thrombectomy cerebral hyperdensity and natural language processing.
A proposed governance framework for GenAI in anatomy education uses four risk tiers (Low, Moderate, High, Restricted/Avoid) under a conservative rule—the highest triggering criterion sets the tier. Permitted uses then pass five control gates: inputs/provenance, anatomical fideli…
Interpretation and reporting of coronary computed tomographic angiography (2026 update): An expert consensus document of the Society of Cardiovascular Computed Tomography (SCCT).
Commercial artificial intelligence for pulmonary nodule assessment cut adjusted median chest CT reporting time from 21.3 to 18.2 minutes (14.6%) and raised report completion by 17% in 39,323 exams; thoracic radiologists saw 25% shorter times, but emergency scans took 7.1% longer.
Radin selected to support VA teleradiology through Valor Network
Increased Imaging Rates Linked to Clinician Self-Interpretation
AI cuts chest CT reporting time
PCCT on par with MRI for diagnostic breast imaging
Koning partners with UberDoc marketplace for breast CT patient access
Photonova Spectra Secures CE Mark Approval, Expanding Global Reach
Lumbar CT Outperforms MRI in Diagnosing Vertebral Bone Quality
Can Photon-Counting CT Have an Impact in Breast Cancer Staging?
A Message from R&E Foundation Board of Trustees Chair, Pamela Woodard, MD
Navigating Uncertainty in Active Surveillance for Prostate Cancer
Asplundh Names Chris Farley COO
C]Pyruvate MRI in Participants with Hepatocellular Carcinoma
Scan.com raised $220 million ($90M equity, $130M debt) to expand its U.S. imaging scheduling platform. Over 900,000 patients have booked scans through its network, and its annualized revenue run rate tops $165 million.
A 43-year-old Washington woman at high risk for breast cancer owed $1,191 after her insurer paid just $13.90 for a preapproved breast MRI, despite a state law mandating coverage, KFF Health News reports.
Radiology Imaging Associates Employer Profile
Standardized reporting in gynecological ultrasonography in India is the need of the hour.
Remote Daytime Emergency Radiologist Job Opening in Remote, Remote
CTisus posts its September 2026 computed tomography (CT) interpretation quiz with video answers and explanations, giving radiologists and trainees a self-assessment for CT scanning skills.
