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…
At 90, Dr. Harold Moskowitz—who interpreted >100,000 X-rays—volunteers weekly to teach UConn medical students and radiology residents, driven by his passion. He co-founded the Society of Breast Imaging.
Remote Swing Shift Radiologist: 1 week on/1 week off OR 1 week on/2 weeks off. Up to $75,000 in hiring incentives for 2-10p ET and $150,000 for 5p-1a ET Job Opening in Nationwide
On serial coronary CT angiography, a random forest model predicted coronary artery calcification progression with AUC 0.81 (95% CI 0.78-0.84), outperforming a traditional model (AUC 0.64, 95% CI 0.59-0.68). Baseline calcium score and plaque burden were key predictors.
A Neuroimaging Biomarker for Hippocampal Network Dysfunction in Alzheimer's Disease.
An ensemble machine learning model integrating clinical and MRI features achieved an area under the curve of at least 0.85 for predicting one-year ASIA Impairment Scale grades after traumatic cervical spinal cord injury in an external test set of 98 patients. The model also pred…
A 2D ResNet model automatically classified anterior talofibular ligament (ATFL) tears on MRI with 92.6% accuracy, detecting abnormal ligaments with 95.0% sensitivity and 93.6% specificity. Performance was comparable to a junior radiologist, but the workflow was faster (3 min vs…
A clinically oriented narrative review proposes an adaptive AI cascading model for Alzheimer’s disease, integrating blood biomarkers and selective imaging escalation to mirror real-world workflows and improve clinical translation.
External validation of knee OA progression DL models drops median AUC from 0.81 to 0.75; surgery prediction models (AUC 0.87) may reflect access to care, not disease, per review of 33 studies.
In a multicenter study, SAGE-derived habitat analysis with RBF-SVM differentiated meningiomas, brain metastases, and gliomas with internal macro-AUC 0.700±0.086. External validation macro-AUC was 0.629. A simpler model using mean features alone gave external macro-AUC 0.849.
Dentin-normalized CBCT radiomics with XGBoost achieved an AUC of 0.78 for differentiating periapical granulomas from radicular cysts, significantly higher than raw radiomics (AUC 0.68, P<.05).
Authors' Reply "Can Large Artificial Intelligence-Based Linguistic Models Help to Obtain Information About Burning Mouth Syndrome?".
An externally validated random forest model predicted radiographic adjacent segment degeneration after short-segment lumbar fusion (AUROC 0.782 internal test, 0.749 external). Five perioperative variables were integrated; preoperative intervertebral space height was the stronges…
In cancer patients with chronic subdural hematoma, middle meningeal artery embolization (MMAE) was associated with lower 90-day mortality vs medical management (15.3% vs 20.9%; HR 0.69) and fewer rescue surgeries when added to surgery (8.8% vs 13.7%; HR 0.63).
Vertebral level vs. kidney border for predicting postnatal low conus on fetal MRI: no significant accuracy difference (p=0.66). Postnatal confirmation in 65% of 34 imaged fetuses.
Parent artery plaque enhancement (PAPE) and aneurysm wall enhancement (AWE) on preprocedural vessel wall MRI independently predicted new DWI-positive microembolic events after endovascular treatment of unruptured intracranial aneurysms, with quantitative parent artery enhancemen…
A CT radiomics-based model achieved a test-set AUC of 0.745 for predicting EGFR mutation in stage IA non-small cell lung cancer; incorporating clinical variables modestly improved performance.
ChatGPT outperformed 4 other LLMs in providing microwave ablation guidance, achieving the highest mean quality score (3.65/4) when rated by radiologists. Claude and DeepSeek followed (3.47), while Gemini (3.00) and Meta (2.95) scored lower, suggesting potential for clinical deci…
A Fuzzy Residual-ShuffleNet deep learning model achieved 94.999% accuracy, 95.899% sensitivity, and 93.889% specificity for breast cancer detection on mammograms in a retrospective study.
An AI model combining ConvNeXtV2 and YOLOv8 with genetic algorithm tuning achieved 78.4% precision, 53.6% recall, and 62.6% AP50 for detecting dental caries on panoramic radiographs, outperforming baseline YOLOv8.
FDA 510(k) clearance for XLITE, a radiology device from Gemss Healthcare Co., Ltd., classified under product code IZL.
Higher pericarotid fat density on CT angiography was linked to worse cognition and functional outcomes in cerebral small vessel disease. Maximum pericarotid fat density independently predicted adverse functional outcomes (p<0.001), with a predictive model AUC of 0.939.
Deep learning achieves favorable results for classifying otitis media via otoscopy and segmenting cochlea/ossicular chain on CT/MRI, but data scarcity for rare diseases and poor stapes segmentation limit clinical use.
A public dataset of 642 non-enhanced CT scans covering six pediatric extracranial germ cell tumor subtypes is now available, with preliminary radiomics models showing utility for AI-based diagnosis.
Four miniature jewels for summer reading: Torrey House Press’ Beyond the Glittering World, The Divination of Salmon, Astonished by Beauty, The Sun Has Shifted As Have I
This educational review details MRI-based oxygen extraction fraction (OEF) mapping techniques. Quantitative BOLD (qBOLD) and quantitative susceptibility mapping (QSM) are foundational, but hybrid QSM-qBOLD methods jointly use magnitude and phase data to improve source separation…
<![CDATA[FDA Expands Approval of Pluvicto in Combination with ARPI for PSMA-Positive mAPMN/S Prostate Cancer]]>
Body Imaging Radiologist ($600-$700K, w/ $1.2M+ earning potential) Job Opening in Clovis, California
Mantle Cell Lymphoma with Bleed in Rectum
A new CT quiz for August 2026 is available on CTisus.com, complete with a video providing answers and explanations. The educational resource, founded by Dr. Elliot K. Fishman, focuses on CT scanning and offers related apps on the Apple App Store.
From Reading to Treating: How Theranostics Is Redefining Radiology
