New Subtraction CT Technique Enhances Detection of Pelvic Fracture Hemorrhages
New Subtraction CT Technique Enhances Detection of Pelvic Fracture Hemorrhages
New Subtraction CT Technique Enhances Detection of Pelvic Fracture Hemorrhages
MIPS QCDR Data Validation Audit
Siemens Healthineers Introduces Accela Radiotherapy Solution
New Subtraction CT Technique Enhances Detection of Pelvic Fracture Hemorrhages
ASTRO 2026 | Stereotactic Radiosurgery Demonstrates Potential in Treating Severe Depression
Prenuvo Introduces Research Network to Analyze Disease Patterns using Whole-Body MRI
My PIER Experience: Perspectives From a Medical Student
Step One of the Diagnosis: The Imaging Order
No Home Program? Medical Student Guide to Virtual Research
A nomogram integrating MRI-based deep learning score, immunoscore, and clinical factors achieved AUCs of 0.922 (training) and 0.914 (test) for predicting 5-year distant metastasis–free survival in 306 rectal cancer patients, enabling risk stratification.
Automated artificial intelligence echocardiography in 62 transthyretin cardiomyopathy patients (178 exams) showed moderate agreement with a human reader (intraclass correlation 0.65, 0.51) and repeatability similar to experienced cardiologists (limits of agreement 3.6 mm, 37 mL).
AI-derived left atrial volume from noncontrast coronary artery calcium CT correlates strongly with echocardiography (r=0.82) but is systematically larger (mean 45.7 vs 32.2 mL/m2; bias 13.5 mL/m2). Agreement was robust across sex, BMI, cardiac phase.
CT-derived periaortic fat phenotypes (fat attenuation index AUC 0.819; fragmentation index 0.716) associated with type B aortic dissection vs controls; aortic volume AUC 0.966 (95% CI 0.941-0.991). Exploratory; tear risk index not standalone.
Letter to the Editor: Cartilage T2 elevation in adolescent athletes-adaptation or early risk?
Virtual planar images derived from SPECT provide comparable heart-to-contralateral lung ratios to standard planar imaging for 99mTc-PYP cardiac amyloidosis, with correlation ≥0.97 and statistical equivalence under attenuation and scatter corrections.
Real-world access to PSMA PET/CT for prostate cancer remains below estimated eligible populations in staging and biochemical recurrence, with wide variation between countries. Barriers include reimbursement, capacity, logistics, and non-standardised referral pathways.
A perspective review argues that despite rising claims of expert-level AI performance and a forming regulatory framework, translating AI tools into reliable clinical value in Indian radiology remains uneven. Success will require more than technical capability—workflow integratio…
AI in obstetric ultrasound divides into assistive, diagnostic, predictive, and prognostic tools. Only assistive and diagnostic tools are verifiable at point of care by the sonographer, while predictive and prognostic tools require caution.
Restriction spectrum imaging in depression patients shows ECT causes right amygdala/hippocampus hindered water increase, free water decrease, and volume increase; volume partially resolves but amygdala extracellular changes persist, suggesting glymphatic flow shifts.
In a clinical brain MRI study of 2,863 patients, greater neighborhood socioeconomic disadvantage was linked to lower brain volumes in cerebellar, frontal, and medial temporal networks in men but not women. Only cerebellar network association remained significant in sex-interacti…
Spine MRI protocols for spontaneous intracranial hypotension (SIH) show substantial heterogeneity in fat-suppression and echo time despite universal 3D T2-weighted imaging use across 9 US expert centers. Ambiguous terminology like 'MR myelography' complicates comparisons.
Radiologist artificial intelligence (AI) adoption hit 91.4%; AI use was tied to 26% lower report turnaround for trauma radiographs and 18% lower for knee magnetic resonance imaging (MRI) (p<0.001). Median latency: 2.06 min, 72% from data routing; 7.2% arrived after finalization.
Federated Learning - Towards Privacy-Preserving, Adaptable and Scalable Artificial Intelligence Integration in Clinical Breast Imaging Workflow.
Patients with a low-risk (compact) necrosis phenotype on post-chemoradiation MRI had nearly three-fold lower hazard of death (HR 0.35, p=0.001), independent of tumor volume, MGMT status, age, and KPS.
In 139 biopsy-referred breasts, tomosynthesis and synthesized-2D AI malignancy scores showed only moderate agreement (κ=0.53) despite similar AUCs (0.728 vs 0.717). Adding either score to a clinical model improved fit but not discrimination meaningfully. Do not downgrade a suspi…
In 319 neonates with perinatal asphyxia, MRI count of enlarged perivascular spaces was independently associated with brain injury severity (p=0.0001) and, in severe cases, later sleep disturbance (r=0.3365, p=0.0061).
TEMPORARY REMOVAL: Association between Technical Component Billing for Medical Imaging and the Likelihood of Radiology Practice Closure.
A new cortex-grounded diffusion model, Cor2Vox, synthesizes realistic brain MRIs by leveraging high-resolution cortical surface priors. It enabled anatomy-guided synthesis, simulation of gray matter atrophy, and harmonization of dementia scans, outperforming baseline methods.
AI-enhanced virtual simulation can complement physical labs and clinicals in radiography education, not replace them. A three-layer model (AI prep, lab practice, supervised placement) is proposed for resource-limited settings. Expert validation critical as AI may be inaccurate.
A deep learning model fusing CNN and Vision Transformer features on MRI achieved external validation AUC 0.810 (95% CI 0.710–0.911) for stratifying high-grade glioma survival after temozolomide, in a multicenter study of 507 patients.