Radiology
Audio Summaries

Associations of Popliteal Artery Wall Morphology at MRI and Knee Osteoarthritis in the Osteoarthritis Initiative

Aug 4, 2026

The authors aimed to investigate the relationship between popliteal artery wall morphology, as measured by MRI, and both symptomatic and radiographic outcomes of knee osteoarthritis (OA). Their findings revealed that localized wall thickening was linked to increased pain and worsening symptoms, while diffuse wall remodeling correlated with symptomatic worsening but showed an inverse relationship with radiographic OA outcomes. This suggests that different aspects of popliteal artery morphology may have distinct implications for knee OA progression.

Radiologically Relevant Clinical History Summarization with Large Language Models: A Multireader Performance Study

Aug 4, 2026

The authors aimed to determine whether large language models (LLMs) could enhance the clinical utility of imaging indications derived from clinical notes, addressing the issue of incomplete clinician-provided histories. Their study found that indications generated by LLMs were significantly more comprehensive and factual than those from referring clinicians, with the proprietary LLM being rated as the most useful for protocoling and interpretation. This suggests that LLMs could improve diagnostic accuracy and workflow efficiency in radiology.

How I Do It: Understanding and Leveraging Generative AI for Chest Radiology Reporting

Aug 4, 2026

The authors aim to explore how generative AI can be effectively utilized in chest radiology reporting, focusing on its potential to enhance diagnostic interpretation and workflow. They provide a practical framework for implementing generative AI, highlighting its benefits, such as improved reporting consistency and trainee education, while also addressing limitations like hallucinations. The article concludes with recommendations for safe integration of AI tools in thoracic radiology, emphasizing the need for context-aware modeling and multi-institutional evaluation.

DBT-DINO: Toward Foundation Model-Based Analysis of Digital Breast Tomosynthesis

Aug 4, 2026

The authors aimed to develop and evaluate a foundation model for digital breast tomosynthesis (DBT), named DBT-DINO, and to assess the impact of domain-specific pretraining on various clinical tasks. The study found that DBT-DINO outperformed the baseline model in breast density classification but showed no significant differences in 5-year breast cancer risk prediction or lesion detection, indicating the need for further refinement in domain-specific pretraining for these tasks.

Comparison of Second-line Hepatic Arterial Infusion Chemotherapy and Tyrosine Kinase Inhibitors in Advanced Hepatocellular Carcinoma

Aug 4, 2026

The authors aimed to compare the clinical outcomes of hepatic arterial infusion chemotherapy (HAIC) and tyrosine kinase inhibitors (TKIs) in patients with advanced hepatocellular carcinoma (HCC) following failure of atezolizumab-bevacizumab treatment. The study found that HAIC resulted in higher response rates and longer progression-free survival compared to TKIs, although overall survival rates were similar between the two treatment options.

Photon-counting CT Extracellular Volume Fraction for Liver Fibrosis Assessment: Validation against MR Elastography and Histopathology

Aug 4, 2026

The authors aimed to evaluate the feasibility of photon-counting CT (PCCT)-derived extracellular volume fraction (ECV) for staging liver fibrosis by comparing it with MR elastography-derived liver stiffness measurement (LSM) and using histopathology as the reference standard. Their findings demonstrated a strong correlation between ECV and LSM, with both methods showing equivalent diagnostic performance for liver fibrosis staging, indicating that PCCT-derived ECV is a clinically viable option for assessing liver fibrosis.

Contrast-enhanced Mammography for Breast Cancer Screening in Women at Increased Risk: Comparison of Prevalence and Incidence Screens

Jul 28, 2026

The authors aimed to compare the performance of contrast-enhanced mammography (CEM) between prevalence and incidence screens in women at increased risk for breast cancer. The study found no significant difference in cancer detection rates between the two types of screens, but incidence CEM demonstrated higher specificity and accuracy compared to prevalence CEM. Overall, CEM remained effective in detecting cancers during both screening types, with consistent performance in identifying malignancies.

Combination of Radiation Therapy and Irreversible Electroporation for Intermediate-Risk Prostate Cancer (RTIRE): A Phase 1 Trial

Jul 28, 2026

The authors aimed to assess the feasibility and safety of combining radiation therapy with irreversible electroporation (RTIRE) for treating intermediate-risk prostate cancer. In a phase 1 trial involving 10 participants, the study found that RTIRE was feasible, with no residual disease detected at 12 months and no severe adverse events reported. Additionally, while some patient-reported outcomes showed temporary changes, they did not indicate a clinically meaningful decline over the study period.

Case 348: Orbital Subperiosteal and Epidural Hematomas Secondary to Sickle Cell Bone Infarcts

Jul 28, 2026

This case report describes a 6-year-old boy with sickle cell disease who presented with symptoms indicative of orbital subperiosteal and epidural hematomas following a recent acute chest syndrome episode. The authors aim to highlight the complications arising from sickle cell bone infarcts, particularly how they can lead to significant neurological and ocular issues. Through imaging studies, the report seeks to illustrate the diagnostic challenges and considerations in managing such cases.

SCAN-MRI: Cardiac MRI-integrated Risk Score for Predicting Cardiovascular Events in Type 2 Diabetes Mellitus

Jul 28, 2026

The authors aimed to develop a cardiac MRI-based risk score, named SCAN-MRI, to predict cardiovascular events in patients with type 2 diabetes mellitus, addressing the limitations of existing models that rely solely on clinical risk factors. The study found that the SCAN-MRI model outperformed established risk models, demonstrating improved predictive performance through the integration of cardiac MRI parameters alongside traditional clinical factors. This new model showed good discrimination and calibration in both training and external test sets, highlighting its potential for better risk assessment in this patient population.