The 2026 Chinese lung cancer guideline adds AI and opportunistic screening in low-dose CT, expands molecular testing, and updates treatment regimens for NSCLC, including perioperative, consolidation immunotherapy, and targeted therapy for EGFR, ALK, KRAS, HER2, and MET alteratio…
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.
A multiphase CT angiography spot sign identified high-risk intracerebral hemorrhage: hematoma expansion occurred in 25/51 (49%) phase-1, 13/28 (46%) phase-2, and 4/9 (44%) phase-3 spots versus 12/71 (17%) with no spot sign.
Stacked deep learning ensembles achieved 0.96 accuracy and 1.00 macro-AUC for periapical lesion detection, with 0.94 sensitivity for lesions <2 mm. Small internal study; external validation needed.
A multiparametric MRI habitat imaging model predicted Ki-67 proliferation index ≥5% in meningiomas with AUCs of 0.929 in training and 0.903–0.914 in two external validation sets, outperforming whole-tumor radiomics.
Author Correction: Classification of fallers in Parkinson's disease through machine learning based feature analysis.
An MRI-based network vulnerability index improved prediction of MCI-to-dementia conversion (HR 1.47, p=0.004) when added to plasma p-tau217/Aβ42 model—supporting a plasma+MRI triage pathway for Alzheimer's.
Adding heterogeneous training data reduced segmentation outliers and improved model generalization for vestibular schwannoma, but cochlea segmentation accuracy declined in the 2023 edition of the crossMoDA challenge.
Meta-analysis of nine studies (2985 patients): MRI deep learning distinguishes triple-negative from other breast cancer subtypes with pooled AUC 0.85 (95% CI 0.81-0.88), sensitivity 0.83, specificity 0.87; certainty of evidence low.
Sonoscape Medical Corp. received FDA 510(k) clearance for its family of digital color Doppler ultrasound systems, including the Marvo, Fenix, and Endo series. The devices were classified under product code IYN and reviewed by the Radiology committee.
Fusion model on multiphasic CT differentiated pediatric hepatoblastoma from focal nodular hyperplasia with AUC 0.980, accuracy 94.2%, sensitivity 94.9%, specificity 92.9% in 121 children. Model assistance improved junior radiologist accuracy from 88.4% to 95.0% (p=0.0078).
The ABR released sample cases, a simulation video, and scoring details for the 2028 Radiology Oral Boards: MRI sequences will be labeled except when sequence ID is tested, and scoring uses 68-72 scale (70 passing) with category average ≥70 required.
Southeast Steel Systems Employer Profile
For PI-RADS 4-5 lesions on bpMRI, targeted biopsy (TB) alone vs combined TB+systematic biopsy: csPCa detection 67.9% vs 69.7% (P=0.373, Kappa 0.958). PCa detection 79.6% vs 82.2% (P=0.136). TB may be an equivalent alternative.
In 395 biopsy-naive men with PI-RADS 3 MRI lesions, transition-zone PSA density had an AUC of 0.927 for clinically significant prostate cancer vs 0.902 for whole-gland PSAD. A TZ-PSAD cutoff of at least 0.331 ng/mL2 could avoid 74.9% of biopsies while missing only 2 cancers.
Hybrid Diagnostic Radiologist - Near Akron, OH - Sign-on and Premium Pay Job Opening in Ohio
Development and external validation of a simple pre-biopsy risk stratification model for clinically significant prostate cancer in men with PI-RADS 3 lesions: a multicenter transperineal biopsy study
<![CDATA[Radiology Roundup of FDA News: August 16 — August 22]]>
Week in Review: AI software testing | Wearable ultrasound | Minnies nominations
Quantitative DCE-MRI predicted 17-week pancreatic cancer chemotherapy response with 98% accuracy after 7 weeks, after phantom correction (extended Tofts model). Earlier CT RECIST identified only 17% responders.
In 500 pregnant women with chronic hypertension, left ventricular hypertrophy (LVH) on early echocardiography was independently associated with adverse maternal outcomes (OR 2.16, 95% CI 1.13-4.13) and fetal outcomes (OR 2.42, 95% CI 1.31-4.45).
A deep-learning model (PointNet++) measured acetabular cup orientation on postoperative CT with MAE 4.48° (anteversion) & 3.89° (inclination), outperforming manual (8.91°/8.70°, p<0.05). Internal validation only.
Automated MRI pipeline using U-Net segmentation and Random Forest classified intracerebral hemorrhage temporal stages with an accuracy of 0.807 in an independent validation cohort of 114 patients.
A new deployer-side readiness instrument (RRI-MI) maps 11 governance domains to a 22-point rubric to help radiology practices operationalize post-market AI oversight—from scanner drift monitoring to version control and human oversight—illustrated with a prostate MRI example.
In 39 intracranial aneurysm cases treated with the Contour Neurovascular System, smaller pre-treatment dome and ostium areas predicted positive outcome (Cliff δ −0.57 and −0.43); device shape changes correlated with oscillatory flow.
A retrospective review of 129 Lung-RADS 0 screening CTs found that radiologists reclassified 26.4% as Lung-RADS 2–4; no cancer diagnoses were delayed. 61.2% of findings resolved or improved.
In a study of 2,129 patients, an X-ray-based multimodal deep learning and radiomics model achieved AUC 0.988 (95% CI 0.975–0.990) for staging vertebral compression fractures.
Three OCD subtypes identified via functional neuroimaging: Limbic-Hyperactive (40%), Fronto-Striatal-Hypoconnected (35%), Global-Disrupted (25%). SVM accuracy 81.5% for subtype assignment; limited generalizability for OCD vs control (AUC 0.567–0.673). Need prospective validation.
CT-derived abdominal fat tissue index (AATI) predicted ICU length of stay (+15.5% per SD) and ventilation duration (+30.8% per SD) in 142 trauma ICU survivors, while BMI did not; routine CT body composition may improve risk assessment.
