ASFNR clinical state-of-practice: For posterior circulation large vessel occlusion stroke, use noncontrast CT to exclude hemorrhage, CT angiography to identify occlusion, and MRI with DWI as reference standard. Imaging scores like PC-ASPECTS help select patients for thrombectomy.
The ACR Appropriateness Criteria 2026 update provides imaging guidelines for suspected or confirmed diffuse lung disease, covering initial evaluation, acute exacerbation, and surveillance.
New expert consensus framework for symptomatic B3 breast lesions: assess concordance, lower excision threshold for discordant/palpable lesions, conservative management for concordant non-atypical lesions, and risk-based surveillance for atypia.
ACR Appropriateness Criteria for indolent lung cancer: For patients with slow-growing subsolid nodules or atypical cysts where surveillance is chosen over intervention, imaging follow-up aims to detect changes that alter management, with modality selection guided by accuracy and…
Most patients want to be told if artificial intelligence is used to generate their imaging reports, a new survey finds. Exact percentages were not provided in the source. Transparent communication about AI could be important for maintaining patient trust.
Continuous PSMA-PET metrics (e.g., SUVmax, tumor volume) improved prognostication for patients with mCRPC compared with binary positive/negative reads, reports RSNA News. (Exact effect sizes not provided in source.)
Implementing a dedicated radiology nurse navigator role at UC Davis Comprehensive Cancer Center improved the diagnostic process for breast cancer patients and providers, reducing delays and streamlining imaging follow-ups (specific metrics not reported in source).
When MRI for painful hip arthroplasty appears normal around the hardware and joint, scrutinize the proximal femoral periosteum. A subtle but key finding is a high-signal rim of circumferential periosteal reaction.
RefleXion Medical submitted a 510(k) application to the FDA for Anchor Point, a tumor tracking technology—potential step toward more precise radiotherapy. (Exact details not reported in source.)
The FDA has cleared a new software platform for Siemens Healthineers' Mammomat mammography devices, adding contrast-enhanced mammography (CEM) image reconstruction and biopsy capabilities directly on the system.
STARMED received U.S. FDA clearance for a radiofrequency ablation (RFA) device intended to treat thyroid nodules, providing a minimally invasive option that may reduce the need for surgery.
Pediatric Radiologist in Spokane, WA Job Opening in Spokane, Washington
Ratio Therapeutics has signed a supply agreement with PanTera for actinium-225 (Ac-225), a radioisotope used in targeted alpha therapy. The deal aims to secure the radiopharmaceutical pipeline for cancer treatment.
Microbot Medical taps Sanmina to manufacture Liberty system
Investigational generative AI models for chest X-rays may improve diagnostic accuracy, streamline AI development, and strengthen radiology quality assurance, though most remain research-stage only.
Diagnostic Imaging published its top 5 radiology content picks for June 2026, covering practice-changing research, technology updates, and clinical insights.
Interviews with nuclear medicine leaders highlight the field's rapid expansion toward precision theranostics, AI-augmented image analysis, and next-generation radiotracers for personalized care.
A remote neuro radiology job based in Texas offers a 7-on/7-off evening shift schedule with premium compensation. The position is fully remote and 100% neuroradiology.
A radiology practice must pay $7M and modify its policies after a radiologist misinterpreted a breast mass as benign, leading to a terminal cancer diagnosis.
Novel invisible data-poisoning attack for medical image AI achieved high invisibility (PSNR 43.46 dB, SSIM 0.9925); holistic defense detected poisoned samples with 95.8% accuracy (F1 >0.95) in a simulation across radiology, ophthalmology, and pathology datasets.
A Random Forest model using diffusion MRI graph metrics (betweenness centrality) and clinical scales (SOPS, K10) discriminated transdiagnostic risk stages in youth (n=243). No accuracy metrics were reported; external validation is needed.
FedFound, a federated foundation model trained on 22,911 structural brain MRI connectomes (ages 0–100), demonstrated superior performance across 9 diagnostic tasks spanning neurodevelopmental, neuropsychiatric, and neurodegenerative disorders (exact metrics not reported).
YOLOv11 outperformed a two-stage ROI-guided ResNet for automated scoliosis detection from spinal radiographs; exact accuracy metrics not reported. Retrospective, single public dataset without external validation.
Adding quantitative metrics to visual amyloid PET reads may improve inter-reader reproducibility, tackling known variability in dementia imaging.
APEX-NET, a deep learning model using non-contrast CT, achieved AUCs of 0.949-0.981 for acute pancreatitis diagnosis and 0.873 for severity prediction, matching senior radiologists in a multi-center study of 3383 patients.
Retrospective study of 47 general surgery patients: ChatGPT-based AI recommendations had 85% at least partial concordance with tumor board decisions, but discorded in 15%, especially on resectability and staging, tending to favor broader surgery.
OpenDicomViewer, a new macOS DICOM viewer built in Swift, used 32% of the memory of Horos (9.6 vs 30.0 GB mean) while opening a 9.3 GB dataset, with a codebase >23 times smaller, potentially easing AI-assisted development after future external validation.
NC-AMRI for HCC surveillance: high interreader agreement among early-to-mid career radiologists. Image quality PABAK 0.81-1.00, observation detection κ 0.74-0.99, AMRI-LI-RADS κ 0.64-0.92. Less experienced readers rated quality higher and detected more observations (retrospectiv…
A narrative review finds digital pre-exam check-in can boost CT/MRI information completeness, cut no-shows, and flag risks earlier — but benefit depends on system integration and patient-centered design, not standalone tools (exact figures not reported in source).
A UK survey of 273 radiographers found limited AI literacy and low confidence to lead AI implementation; most preferred operational roles after system deployment. Women and those with little AI exposure reported significantly lower confidence (exact figure not reported in source…
Expert review proposes a PRISM-RE-AIM implementation framework for starting IGPO programs, emphasizing institutional readiness, stakeholder engagement, and metrics like technical success and adverse events.
In HCC ≤5 cm bridged to transplant, radiation segmentectomy (Y90) and TACE+MWA showed equivalent tumor necrosis on explant: mean 79% vs 84% (p=0.48), complete necrosis 45% vs 62% (p=0.14). Tumor size ≤3 cm predicted CPN (OR 18.3, 95% CI 2.1–156.5).
Multi-omics deep learning (CT+genomics) predicted benefit from anti-EGFR therapy in RAS wild-type colorectal liver metastases (AUC 0.86), with HR 17.9 for sensitive cases and better PFS (9 vs 5 months). External validation cohort small.
A combined model integrating hippocampal subfield radiomics, whole-brain GMV, and clinical data achieved AUC 0.875 for predicting in-hospital aggression in schizophrenia, outperforming standalone imaging models (AUCs 0.681–0.729). Retrospective single-center study.
