Radiology
Audio Summaries

Thrombectomy in Posterior Circulation Tandem Occlusions: Multicenter Comparative Analysis of Procedural Techniques and Predictors of Clinical Outcomes

Jun 23, 2026

The authors aimed to evaluate the outcomes of thrombectomy in patients with posterior circulation tandem occlusions (TOs) across multiple centers. They found that thrombectomy was feasible and effective, with successful reperfusion achieved in 88.5% of patients and favorable functional outcomes associated with a distal-first strategy. Additionally, factors such as hyperglycemia, hypertension, and higher baseline NIHSS scores predicted unfavorable outcomes, while successful reperfusion and prestroke antithrombotic use were linked to better outcomes.

Standardized Knee Meniscus MRI Reporting: An Interdisciplinary Delphi Consensus

Jun 23, 2026

The authors aimed to establish standardized interdisciplinary guidelines for reporting knee MRI findings related to meniscus conditions through a Delphi consensus process involving 33 panelists from 23 institutions. Over three rounds, the panelists achieved consensus on various topics, including MRI criteria for tears, tear descriptors, and assessment of postoperative menisci, highlighting significant agreement on critical reporting elements. This consensus aims to improve the consistency and clarity of knee meniscus MRI reporting across different medical specialties.

Longitudinal Analysis of Changes in Deep Learning Image-based Breast Cancer Risk Scores over Time

Jun 23, 2026

The authors aimed to investigate whether deep learning image-based breast cancer risk scores change over time and if these trajectories differ between women who develop breast cancer and those who remain cancer-free. Their study found that risk scores significantly increased in women who developed cancer, while remaining stable in cancer-free women, suggesting that these AI-based scores could serve as dynamic biomarkers for risk-adaptive screening and prevention strategies.

MR Lymphangiography for Diagnosis of Lower Extremity Lymphedema: Suppressing Venous Signal Interference Using Deep Learning

Jun 23, 2026

The authors aimed to develop and evaluate LympClear, a deep learning-based method for suppressing venous signal interference in MR lymphangiography to improve the diagnosis of lower extremity lymphedema (LEL). Their results demonstrated that LympClear significantly enhanced image quality, increased the visibility of lymphatic reflux, and improved diagnostic efficiency compared to standard imaging techniques. Overall, the study indicates that LympClear is an effective tool for enhancing the diagnostic accuracy of MR lymphangiography in LEL patients.

Extending the PREVENT Equations with Cardiac MRI: Prediction of 10-year Heart Failure Risk

Jun 23, 2026

The authors aimed to determine whether incorporating multidimensional cardiac MRI parameters into the existing PREVENT equations could enhance the prediction of 10-year heart failure risk. Their analysis of data from 39,069 participants revealed that the integration of 16 cardiac MRI parameters significantly improved risk prediction, with notable differences in contributions between sexes. This study underscores the potential of cardiac MRI in refining heart failure risk assessments.

Ten-year Longitudinal Relationship between Spinal Degenerative Lesions in Axial Spondyloarthritis at MRI and Radiography in the DESIR Cohort

Jun 23, 2026

The authors aimed to investigate the longitudinal relationships between spinal degenerative lesions in patients with axial spondyloarthritis (axSpA) over a 10-year period, utilizing MRI and radiographic data. Their findings revealed significant temporal associations between various degenerative lesions, indicating a progressive nature of spinal degeneration in this cohort. Notably, MRI lesions were linked to the emergence of corresponding degenerative changes on subsequent radiographs, underscoring the interconnectedness of these imaging findings.

Deep Learning Detection of Direct and Indirect Imaging Findings Associated with Pancreatic Cancer at Contrast-enhanced and Noncontrast CT

Jun 16, 2026

The authors aimed to develop and evaluate deep learning (DL) models for detecting both direct and indirect imaging findings associated with pancreatic cancer (PC) on noncontrast and contrast-enhanced CT images. The study found that these DL models demonstrated diagnostic performance comparable to or better than that of experienced physicians, particularly in identifying smaller pancreatic cancers. Overall, the models effectively detected key imaging findings and diagnosed PC, suggesting their potential utility in clinical practice.

Genicular Artery Embolization Using Rapidly Resorbable Gelatin-based Microspheres for Osteoarthritis-related Knee Pain

Jun 16, 2026

This study investigates the safety and clinical outcomes of genicular artery embolization (GAE) using rapidly resorbable gelatin-based microspheres (RRGMs) in patients with osteoarthritis-related knee pain who have not responded to conservative treatments. The results indicate that GAE with RRGMs is safe, with no severe adverse events, and leads to significant pain reduction and improvement in knee function over a 12-month follow-up period.

Hybrid Cardiac ⁶⁸Ga-FAPI-4 PET/MRI in Dilated Cardiomyopathy: A Feasibility and Pilot Study

Jun 16, 2026

This study aimed to evaluate the feasibility and performance of a hybrid cardiac ^68Ga-FAPI-4 PET/MRI protocol in patients with nonischemic dilated cardiomyopathy (DCM). The results demonstrated that the protocol effectively distinguished DCM patients from healthy controls and oncology patients, with significant correlations between FAPI uptake and various cardiac MRI metrics, suggesting its potential for simultaneous molecular and structural imaging in myocardial diseases.

Zero-shot Thoracic Oncologic History Generation for Radiologists Using Retrieval-augmented Large Language Model Pipeline

Jun 16, 2026

The authors aimed to develop and evaluate a zero-shot large language model (LLM) pipeline for generating structured oncologic histories to improve efficiency in gathering clinical summaries for thoracic oncology patients. Their study found that the GPT-5-mini model achieved the highest completeness and accuracy in summarization, while also demonstrating significant time savings compared to traditional manual methods, potentially leading to increased revenue for radiologists. Overall, the LLM pipeline effectively summarizes oncologic history without the need for manual fine-tuning or information retrieval.