Journal highlights
The following are highlights from the current issues of RSNA’s peer-reviewed journals.
AI-Supported Review Helps Standardize Screening Mammography
Despite evidence that breast imaging specialists outperform general radiologists, about 70% of screening mammograms are interpreted by generalists due to specialist shortages. AI-supported workflows may help address variability and standardize screening mammography performance.
In a new Radiology article, Matthew McCabe, PhD, DeepHealth in Somerville, MA, and colleagues evaluated whether a multistage AI-driven workflow could improve screening performance of generalists and specialists.
The prospective study included 60 generalists and 35 fellowship-trained specialists who interpreted 577,742 mammograms. The AI workflow incorporated a computer-aided detection and diagnosis tool that routed AI-identified suspicious examinations not initially recalled for additional expert review.
With AI, generalists’ cancer detection rate increased from 3.76 to 4.99 cancers per 1,000 mammograms, and positive predictive value increased from 3.38% to 3.89% despite a higher recall rate (9.06% to 10.40%). Specialists’ detection rates were unchanged, and generalists using AI performed similarly to specialists using AI.
“The benefit of improved positive predictive value of recalls indicates a favorable shift in the screening process despite the observed increase in recall rate,” the authors conclude.
Read the full article, “Closing the Performance Gap between Generalists and Breast Imaging Specialists Using a Nationally Deployed AI Workflow for Screening Mammography.”
Follow the Radiology editor on X @RadiologyEditor.
Example mammograms from ProFound Pro 2.x (DeepHealth) output show localized intermediate- (orange) and high- (red) suspicion bounding boxes in craniocaudal and mediolateral oblique views. Both examinations were labeled as highly suspicious but not recalled by the initial radiologist, triggering the Safeguard Review. Upon review by a breast imaging expert, both examinations were recalled, and cancer was diagnosed (cancer location is indicated by arrows).
Complete legend details can be found at https://doi.org/10.1148/radiol.252005 ©RSNA 2026
Imaging Clues of Early Pancreatic Cancer
Pancreatic cancer is often diagnosed at an advanced stage and carries a poor prognosis, making earlier detection critical to improve outcomes.
In an article published in RadioGraphics, Shintaro Kano, MD, University of Tokyo, and colleagues provide a comprehensive overview of the clinical implications, diagnostic workup and imaging findings of early pancreatic cancer, with histopathologic correlation to support personalized surveillance.
The authors defined early pancreatic cancer as high-grade dysplastic precursor lesions and invasive carcinomas measuring 2 cm or smaller that remain confined to the pancreas. Direct imaging findings include a solid lesion or an enhancing mural nodule within a cystic lesion. Equally important are indirect findings, such as focal pancreatic parenchymal atrophy, focal faint parenchymal enhancement, main pancreatic duct strictures and focal side-branch dilatation or retention cysts. Notably, these subtle changes may precede visible tumor development by months to years, providing a valuable window for intervention.
“Radiologists should recognize subtle imaging features that may precede overt tumor development, aiding effective multidisciplinary decision making and timely management,” the authors write.
Read the full article, “Early Pancreatic Cancer: Clinical Implications, Workup, and Imaging Findings with Histopathologic Correlation for Personalized Surveillance,” and the invited commentary, and the invited commentary, "Early Detection of Pancreatic Cancer: Are We Up for the Challenge?” This article is also available for CME on EdCentral.
Follow the RadioGraphics editor on X @RadG_Editor.
High-grade pancreatic intraepithelial neoplasia (PanIN) in a 73-year-old woman who had a history of pancreatic cancer in her mother and was found to have main pancreatic duct (MPD) dilatation during a routine health checkup. (A) Coronal MIP image from 3D MRCP shows an MPD stricture (arrow), with upstream duct dilatation (arrowhead). (B) Axial contrast-enhanced pancreatic phase dynamic CT image shows focal pancreatic parenchymal atrophy (arrowheads) surrounding the dilated duct. (C) Axial contrast-enhanced delayed phase dynamic CT image shows focal faint parenchymal enhancement (arrow). Based on these imaging findings, early pancreatic cancer was suspected. Surgical pathologic analysis confirmed high-grade PanIN extending approximately 45 mm along the MPD.
https://doi.org/10.1148/rg.250085 ©RSNA 2026
Dual-Phase MRA Improves Heart Disease Imaging
High-resolution MR angiography (MRA) plays a critical role in the evaluation and management of pediatric patients with complex congenital heart disease (CHD), where precise visualization and measurement of cardiovascular anatomy are essential for diagnosis and procedural planning. However, image quality and vessel dimensions can vary depending on the cardiac phase, and arrhythmia or high heart rate variability may limit diagnostic adequacy.
In a recent article published in Radiology: Cardiothoracic Imaging, Lucia D. Beissel, MD, University Hospital Bonn, Germany, and colleagues evaluated a novel free-breathing dual-phase three-dimensional modified Dixon steady-state MRA technique in children with complex CHD. The authors demonstrate that dual phase imaging enables phase-dependent assessment of vessel size and image quality within a single acquisition, with systolic imaging improving visualization of select structures and mitigating non diagnostic exams in challenging cases.
“Dual-phase acquisition may support more reliable preinterventional planning and reduce the likelihood of nondiagnostic examinations, particularly in pediatric patients with elevated or irregular heart rates,” the authors explain. Read the full article, “Dual-Phase 3D Modified Dixon Steady-State MR Angiography in Pediatric Complex Congenital Heart Disease.”Journal Peer Reviewers Needed
Contribute your expertise by volunteering as a peer reviewer for one of RSNA’s prestigious journals. Refine your critical appraisal skills while helping advance high-quality, clinically relevant radiology research.
You’ll gain early access to cutting-edge science, build your professional reputation and expand your network. Peer reviewers are also eligible to earn CME credit. Learn more about peer review and other involvement opportunities now.
.