Invited Commentary: Pancreatic Cancer Screening and Early Detection: Current Perspectives and Future Opportunity.
Automated AI analysis of routine CTs can detect low bone density, hepatic steatosis, sarcopenia, adiposity, and vascular calcifications—conditions often overlooked that contribute to fracture risk, cardiometabolic disease, and frailty. Adoption remains limited by workflow, stand…
Menkes Disease is a disease.
Mammography is safe but less sensitive in pregnancy/lactation. Ultrasound is first-line for symptoms but its screening role is uncertain. MRI is contraindicated during pregnancy, may be used in lactation for high-risk. Balance diagnostic performance with maternal and fetal safet…
An RSNA review outlines evidence-based ultrasound (US) screening for high-risk infants: preterm, breech, syndromic, tumor predisposition. Key: risk-based selection, standard protocols, radiologist expertise in disease patterns. Enables early intervention.
Invited Commentary: Importance of Breast Cancer Screening in Pregnant and Lactating Patients.
Bayesian reasoning is a practical approach for radiology interpretation and reporting.
About one-third of lung screening CTs reveal non-cancer abnormalities. Radiologists can use this practical guide to report and manage opportunistic findings—emphysema, coronary calcium, bone density, abdominal lesions—adding substantial preventive value.
Invited Commentary: Measured, Not Yet Managed: Moving Toward Comprehensive Lung Screening.
U-Net model trained on 462 CyberKnife prostate plans predicted mean doses with absolute error of 0.63 Gy (rectum) and 1.04 Gy (bladder) on a 93-patient test set, showing promise for automating treatment planning.
Review of 98 DTI-based AI studies for Alzheimer's (2010–2026): algorithm choice tied to sample size (SVM for small, DL for large), external validation degraded performance, and certain tracts (hippocampal, callosal, fornix, cingulum) were robust.
AAP updates pediatric UTI guideline (8 days–5 y): 8 Key Action Statements (2 strong, 6 conditional) and 4 Good Practice Statements cover diagnosis, antibiotic treatment, imaging for structural abnormalities/vesicoureteral reflux, and prevention. GRADE certainty low to very low.
Multimodal AI model combining ultrasound radiomics, hemodynamics, and clinical data achieved AUC 0.95 (95% CI 0.91-0.99) and 100% sensitivity for classifying BI-RADS 4a breast nodules, potentially avoiding 32.4% of unnecessary biopsies.
Artificial intelligence for mucus plug assessment: toward a new generation of airway imaging biomarkers.
A review details how AI tools are being implemented across neuroradiology—from scheduling and acquisition to lesion detection and outcome prediction—showing promising technical performance but lacking robust real-world evidence.
Developmental Dysplasia of the Hip: From Newborn Screening to Adult Presentation and Management.
CT colonography, an accurate CRC screening test now reimbursed by Medicare, has sensitivity comparable to colonoscopy for significant polyps. It enables polypectomy of large polyps only, reducing overdiagnosis and perforation risk. Underutilized due to practice barriers.
Invited Commentary: Advancing Screening CT Colonography: Timely Update and Call to Action.
A client-adaptive vision-language gatekeeper reduced radiologist annotation time by ~30% (1.14 hours) in a breast-density federated learning task, retaining 84.4% of in-distribution images and achieving artifact-discrimination AUC 0.85.
A viewpoint defines 'refined exclusion' — when medical AI performs well overall but fails in underrepresented groups — and proposes lifecycle decision gates (proceed, enrich, restrict, pause) to shift focus from average performance to demonstrable safety for all populations.
SLMs (≤4B parameters) showed relative task scores from 0.30 to 2.36 across 9 studies, but the evidence is too heterogeneous to compare with larger models, and safety metrics like calibration remain unreported.
Refractory Chronic Lymphocytic Leukemia Successfully Bridged to CAR T-Cell Therapy Using Splenic Radioembolization.
CT radiomics offered no improvement over volume dynamics for predicting early lesion response in metastatic melanoma on immunotherapy, with both models achieving similar AUCs internally (0.86 vs 0.84) and externally (0.68).
An expert-committee AI framework using quantitative CT features achieved AUCs of 0.918 for AAH/AIS, 0.860 for minimally invasive adenocarcinoma, and 0.913 for invasive adenocarcinoma and inflammatory nodules in a 491-nodule internal test; external validation showed consistent pe…
A dual-teacher semi-supervised model using only 30% pixel labels recovered 81.5% of fully supervised segmentation (Dice 0.626) and calcification IoU 0.700. On INbreast, its classifier had AUC 0.872 and sensitivity 0.837 vs senior radiologists' 0.803 (p<0.05).
From Zero to Interventional Oncology: Bringing Specialist Care Closer to Home in a Tertiary Interventional Oncology-Naïve Hospital.
Beyond DXA bone density, single-component tools like trabecular bone score, quantitative ultrasound, and AI-driven analysis of routine X-rays offer complementary insights into bone strength and fracture risk, enabling opportunistic screening.
Tau PET signal in the entorhinal cortex predicted progression from mild cognitive impairment to Alzheimer's disease only in patients whose entorhinal cortex was close to the tentorial incisura, suggesting chronic mechanical stress may play a role (p=0.036, n=47).
AI translation in MSK imaging is selective: narrow, verifiable radiographic tasks (fracture, morphometry) dominate commercial products; OA MRI lags. Clinical value depends on relevant endpoints, workflow fit, and human-AI interaction—not just benchmark performance, authors argue.
Beyond Bosniak Class: Nodule Size Refines Cystic Renal Mass Risk.