Journal of the American College of Radiology : JACR
Audio Summaries

Paid Family and Medical Leave Policies in Academic Radiology and Radiation Oncology Departments: A SCARD and SCAROP Survey

May 10, 2026

The authors aimed to characterize the availability and implementation of paid family and medical leave (PFML) policies in academic radiology and radiation oncology departments across the United States. Their survey revealed that while PFML policies are widely available and viewed as important for recruitment and retention, actual utilization remains low, and concerns about clinical coverage and clarity regarding promotion clock extensions persist. These findings indicate a gap between the existence of PFML policies and their effective application in practice.

Comparative Analysis of Large Language Models Performance in Appropriate Diagnostic Imaging Modality Selection

May 10, 2026

This study aimed to compare the performance of seven large language models (LLMs) in selecting appropriate diagnostic imaging modalities based on ACR guidelines. Using 50 clinical vignettes, the authors assessed the models' accuracy, clinical reasoning, and citation quality, finding that while all models provided reasonable imaging recommendations, there was significant variability in citation validity, particularly with Google's Gemini models. The results highlight the importance of aligning clinical reasoning with reliable sources before implementing LLMs in clinical practice.

Pre-test Messaging to Reduce Cancellations in Nuclear Medicine: Evaluation in Cardiac Stress Testing and FDG PET Scheduling

May 10, 2026

The authors aimed to evaluate the effectiveness of pre-procedure messaging in reducing appointment cancellations, rescheduling, and delays for cardiac stress testing and FDG PET at a U.S. tertiary hospital. Their findings indicated that pre-procedure messaging significantly decreased cancellations and rescheduling rates, particularly for FDG PET appointments, and also reduced appointment delays, suggesting that enhanced communication can improve scheduling efficiency and patient adherence in nuclear medicine services.

Pediatric MRI Safety: Frequency, Types, and Severity of Zone IV MRI Safety Events at Five U.S. Children's Hospitals

May 10, 2026

The authors aimed to assess the frequency, types, and severity of MRI safety events occurring in Zone IV at five U.S. children's hospitals. Over a six-year period, they identified 146 safety events from approximately 540,987 MRI examinations, with a notable percentage involving projectiles and thermal injuries, and highlighted that many incidents resulted from non-compliance with safety protocols. The findings underscore the importance of ongoing safety improvements in pediatric MRI practices despite the low incidence of serious harm.

Survival in patients diagnosed with lung cancer after low-dose CT screening

May 10, 2026

The authors aimed to evaluate survival and mortality rates in lung cancer patients diagnosed after low-dose CT screening and to assess how these rates correlate with patient and cancer characteristics. The study found that while screen-detected lung cancers were associated with lower mortality, the overall mortality rates in their cohort were higher than those reported in previous research trials, indicating potential differences in outcomes between clinical practice and trial settings.

ACR's Assess-AI: A Registry for Real-World Performance Monitoring of Clinical Imaging AI

May 1, 2026

The authors aim to describe the design and functionality of the ACR's Assess-AI registry, which is established to monitor the real-world performance of clinical imaging AI models. This registry allows facilities to submit AI outputs and associated data, enabling the evaluation of AI performance through standardized metrics and interactive analytics. By addressing gaps in post-deployment monitoring, Assess-AI seeks to enhance the governance and clinical utility of imaging AI technologies.

Shifts in Urbanicity-Based Access to Radiotracers for Neuroendocrine Tumors

Apr 30, 2026

The authors aimed to assess how the introduction of newer radiotracers, specifically ^68Ga-DOTATATE/DOTATOC and ^64Cu-DOTATATE, affected geographic access to radiotracers for patients with neuroendocrine tumors (NETs) compared to the traditional ^111In-pentetreotide. Their findings indicate that while patients from small or rural areas initially faced greater travel distances for gallium-based radiotracers, the introduction of copper-based radiotracers improved access and reduced travel distances, particularly for those in less urbanized regions. Overall, the study highlights the impact of radiotracer availability on patient access and treatment options for NETs.

Geographic Analysis of Digital Breast Tomosynthesis Availability in High- and Low-Deprivation Areas

Apr 19, 2026

The authors aimed to evaluate the relationship between neighborhood deprivation and the availability of digital breast tomosynthesis (DBT) at FDA-accredited breast imaging facilities. Their findings indicate that while most facilities provide DBT regardless of neighborhood deprivation levels, significant disparities exist in weekend service availability and online scheduling, suggesting potential areas for improvement to enhance access for underserved populations.

Industry-Sponsored Research Payments in Radiology: A National Open Payments Analysis

Apr 18, 2026

This study aimed to assess trends in industry-sponsored research payments to radiologists from 2019 to 2024, utilizing data from the CMS Open Payments Database. The findings revealed that radiologists received only 1.1% of total industry research payments in 2024, with a notable decline in their proportional share over time and significant funding concentrated among a small percentage of radiologists. The authors suggest that professional societies and academic departments should develop strategies to enhance radiology's competitiveness for industry-sponsored research funding, particularly in teaching hospitals.

Vibe Coding in Radiology: Claude Code and ChatGPT Codex Leading to the Rise of the Radiologist-Developer

Apr 18, 2026

The authors explore the potential of large language model-based coding agents, such as Claude Code and ChatGPT Codex, to empower radiologists to become developers by simplifying the coding process. They aim to highlight how these tools can facilitate the creation of innovative solutions in radiology education, clinical practice, and research, while emphasizing the need for ethical oversight in their application.