Deep Learning Framework Automates Lenke Classification on Multi-View Spine Radiographs
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.
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…
DenseNet121 model classified cochlear implant electrode position on X-rays with an AUC of 0.932 (95% CI 0.836–1.000) and 98.6% specificity, with interpretable attention maps.
In 530 acute ischemic stroke patients, the automated DWI biomarker SCR-U1.8 logistic regression model achieved 0.990 accuracy and 0.998 AUC for predicting malignant cerebral edema, though only 12 events occurred; findings are exploratory and require external validation.
A study in 972 men externally validated the PROMOD biopsy strategy. PI-RADS 4-5 or PI-RADS 3 with a PSA density >0.15 avoided 18-20% of biopsies while missing 5-8% of clinically significant cancers. A more conservative approach (PI-RADS 4-5 or PI-RADS 3 with PSA density >0.10) a…
A multimodal model integrating preoperative CT and biopsy pathology predicted high-risk recurrence in stage IA NSCLC with AUC 0.76 (95% CI 0.58-0.90). High-risk patients had 2-year DFS 78% vs 99%.
AI-assisted echo measurements of aortic root and LVOT showed comparable CMR agreement to human readers (ICC 0.70-0.77 vs 0.51-0.85), but 36.7% of LVOT, 20% of STJ, and 11.8% of SoV had clinically significant differences (≥2mm/≥4mm). Physician oversight remains necessary.
In 80 patients with thyroid nodules, combined ACR TI-RADS and Bethesda classification: concordant high-risk designations were uniformly malignant. TI-RADS sensitivity 77.8%, specificity 90.3%, AUC 0.91; Bethesda NPV 95.4%.
Union complains about GE HealthCare labor dispute
Early CTA leads to improved stroke outcomes for patients with BCVI
Patient-friendly radiology reports improve family comprehension.
Your First Radiology Contract: 3 Things to Review
Remote/Hybrid Overnight Neuro or Body Radiologist - Partnership Track Job Opening in Remote, Remote
AI Challenge Knee MRI
MARS Bioimaging installs first portable photon-counting CT system
Circle CVI updates cardiovascular imaging platform
Breast Imaging Training in Indonesia
<![CDATA[Study Finds 33 Percent Increase in Radiology Turnaround Times for Medicare Beneficiaries in 2024]]>
Most histotripsy procedures used for metastatic disease treatment
It's Better to Drive the Bus Than to Ride in It. #ImagingAI
Impact of CTA Timing on Outcomes in Blunt Cerebrovascular Injuries
Northside Radiology Associates, a Radiology Partners affiliate with 70+ physicians, signs a multi-year in-network agreement with Anthem BCBS of Georgia, ensuring continued in-network access to diagnostic and interventional imaging for members.
ACR released a handout helping patients understand AI-generated radiology report summaries, advising they discuss unclear results with their radiologist or provider. The resource aims to improve patient understanding and can be incorporated into workflows.
Imaging turnaround times continued to climb in 2024
Photon-counting CT is comparable with MR elastography for liver fibrosis.
PET imaging reveals tau pathology in 65% and amyloid in 35% of patients with psychosis beginning after age 40, vs 15% and 2% in controls, suggesting neurodegenerative disease drives these symptoms.
A new multidisciplinary clinical educator team across a large medical imaging network boosted self-reported confidence in first-trimester communication, from 2.7 to 3.5 on a 4-point scale, and increased education reach. Early results support the workforce model.
An automated AI TAVR planner showed excellent agreement with semi-automated software for aortic annulus area (ICC 0.976, 95% CI 0.966-0.984) and reduced planning time from 9.55 to 1.03 minutes per case (p=0.001).
BI-RADS v2025 brings mammographic density alignment with Mammography Quality Standards Act, nonmass lesions on ultrasound, revised MRI enhancing lesion terminology, and first dedicated contrast-enhanced mammography section, harmonizing lexicon across modalities.
An MRI radiomics model predicted loss of photoreceptorness (a tumor differentiation marker) in retinoblastoma with an AUC of 0.81 in an independent testing cohort, providing a potential non-invasive biomarker for a cancer where biopsy is contraindicated.
To Fellow or Not to Fellow?
Mammography-Based Deep Learning Model Better Than Breast Density at Predicting Breast Cancer Risk.