Increased Emergency Department Medical Imaging: Association with Short-Term Exposures to Ambient Heat and Particulate Air Pollution Open Access 

Radiology, published 11/19/24 | Estimated Read Time: 10 minutes

10-year retrospective analysis of over 1.6 million emergency department imaging studies found short-term exposures to higher ambient temperatures (10°C rise) and particulate air pollution (10 μg/m3 rise in PM2.5) were associated with 5.1% and 4.0% increases in overall imaging utilization, respectively. Heat exposure days (>20°C) and air pollution days (PM2.5 >12 μg/m3) were linked to excess risks of 5.5 and 6.4 additional imaging studies per 1 million people per day, highlighting the need for radiology departments to prepare for potential surges during extreme environmental conditions. 

Incidental Detection of Previously Unknown Strokes on Head CT Examinations: An Untapped Opportunity for Secondary Prevention

Journal of the American College of Radiology, published 11/18/24 | Estimated Read Time: 15 minutes

Retrospective analysis of 21,985 head CT exams found that 4.0% (n=869) detected old strokes, with 43% (n=372) of these being previously unknown to patients and clinicians - a prevalence of 1.7% across all head CTs. Factors associated with unknown strokes included older patient age and smaller stroke size, highlighting the potential to leverage these incidental findings to implement secondary prevention strategies and improve long-term outcomes for affected patients.

CT-Defined Coronary Artery Calcification as a Prognostic Marker for Overall Survival in Lung Cancer: A Systematic Review and Meta-analysis Open Access 

Academic Radiology, published 11/18/24 | Estimated Read Time: 15 minutes

Comprehensive meta-analysis of over 2,200 lung cancer patients found that increased coronary artery calcification on CT scans was significantly associated with worse overall survival, with a pooled hazard ratio of 1.56 in multivariable analysis. The authors also reported a nearly 2-fold higher odds of major cardiovascular events in patients with coronary calcifications, highlighting the importance of assessing this imaging biomarker as a prognostic indicator in lung cancer. 

Evaluating tibial rotation in recurrent patellar dislocation with four-dimensional computed tomography 

Skeletal Radiology, published 11/16/24 | Estimated Read Time: 20 minutes

This study utilized 4D CT and 3D-3D registration to evaluate 3D knee kinematics in 10 patients with recurrent patellar dislocation and 7 controls. Patients with dislocation exhibited significantly greater external tibial rotation during early knee extension (P<0.05 at 0-10° flexion) compared to a 32% faster onset of external rotation compared to the control group. Additionally, patients showed increased patellar lateral tilt during extension (P<0.01 at 0-20° flexion). These quantitative 3D insights into abnormal tibiofemoral and patellofemoral kinematics in patellar instability could inform surgical planning and improve patient outcomes.

Extremely dense breasts: A comprehensive review of increased cancer risk and supplementary screening methods

European Journal of Radiology, published 11/15/24 | Estimated Read Time: 15 minutes

Comprehensive review examines the latest evidence on supplemental screening methods for patients with extremely dense breasts, including digital breast tomosynthesis, ultrasound, MRI, and contrast-enhanced mammography. While each modality offers unique benefits and limitations, the review highlights that contrast-enhanced mammography demonstrated an 8.7 per 1,000 incremental cancer detection rate in women with extremely dense breasts, comparable to the performance of breast MRI but at a lower cost. 

The Delay of Recanalisation of Acutely Thrombosed Dialysis Arteriovenous Access Until the Next Workday has No Negative Impact on Clinical Outcome Open Access 

Cardiovascular and Interventional Radiology, published 11/18/24 | Estimated Read Time: 15 minutes

Study of 355 thrombosed dialysis access procedures challenges the necessity of weekend interventions, finding no significant difference in technical success (96.6% vs 89.1%) or clinical success (88.1% vs 84.6%) between patients treated after weekend delays versus immediate weekday treatment. While delayed treatment led to higher rates of temporary central venous catheter placement (42.3% vs 18.5%, p<0.001) and longer hospital stays (4.0 vs 2.0 days, p<0.001), there were no additional complications during the waiting period, suggesting that postponing treatment to regular workdays may be a viable option for many centers without 24/7 interventional coverage.

Embolization with gelatin foam in the management of vascularized retained products of conception: a multicenter study by the French Society of Cardiovascular Imaging

European Radiology, published 11/19/24 | Estimated Read Time: 15 minutes

Large retrospective multicenter study (n=224) demonstrates that uterine artery embolization (UAE) using gelatin foam achieves a 94.2% clinical success rate for hemorrhagic vascularized retained products of conception (vRPOC), with minimal complications (0.8% major complications). The presence of hypertrophic uterine artery (OR=0.6, p=0.045) and use of gelatin foam torpedoes (OR=0.57, p=0.001) were associated with higher risk of persistent vRPOC on imaging follow-up, suggesting the importance of technique selection and anatomical considerations in treatment planning.

Interventional Radiology Management of Bone Metastasis Pain: Strategies and Techniques Open Access 

Cardiovascular and Interventional Radiology, published 11/19/24 | Estimated Read Time: 25 minutes

This review addresses the critical challenge of bone metastasis pain management, highlighting IR solutions for the 40-50% of cancer patients who remain undertreated with conventional therapies. The authors present a decision-tree approach for selecting between six major IR techniques (vertebral augmentation, percutaneous osteosynthesis, tumoral ablation, electrochemotherapy, intra-arterial therapies, and percutaneous neurolysis), with reported success rates including 90% pain reduction after tumoral embolization and significant VAS score improvements from 6.9 to 2.7 after electrochemotherapy for bone metastases. 

Interventional Oncology Meets Immuno-oncology: Combination Therapies for Hepatocellular Carcinoma

Radiology, published 11/19/24 | Estimated Read Time: 30 minutes

Groundbreaking review examines how combining locoregional therapies with immunotherapy could transform hepatocellular carcinoma treatment, with the EMERALD-1 trial showing significantly improved progression-free survival (15.0 vs 8.2 months, p<0.05) when combining TACE with immunotherapy plus VEGF inhibition. While current immunotherapy response rates remain limited at 20-30% due to most HCCs (>70%) exhibiting a "cold" immunological phenotype, over 50 ongoing clinical trials are investigating these promising combinations, though the IMbrave050 trial revealed that initial benefits of adjuvant immunotherapy after ablation were not sustained long-term (33.2 vs 36.0 months, HR 0.90 [95% CI: 0.72, 1.12]).

A Meta-Analysis and Meta-Regression of Embolisation Outcomes of Pulmonary Arteriovenous Malformations

Cardiovascular and Interventional Radiology, published 11/18/24 | Estimated Read Time: 25 minutes

Meta-analysis of 44 studies (1,865 patients) reveals that plug embolization achieves significantly higher success rates than coils for treating pulmonary arteriovenous malformations (92-94% vs 86%, p=0.001), with an overall treatment success rate of 86% (95% CI: 84-89%) and impressive 99% immediate technical success rate. Researchers found that younger age (p=0.041), simple PAVM morphology (p=0.020), and plug embolization (p=0.001) were associated with better outcomes. 

Impact of Deep Learning-Based Computer-Aided Detection and Electronic Notification System for Pneumothorax on Time to Treatment: Clinical Implementation Open Access 

Journal of the American College of Radiology, published 11/18/24 | Estimated Read Time: 10 minutes

Analysis of 603,028 chest radiographs found that implementing deep learning-based computer-aided detection with electronic notification system reduced time to oxygen supplementation for pneumothorax patients by 143.8 minutes (95% CI: -277.8 to -9.9, p=.035) compared to CAD-only usage, though this lost statistical significance after Bonferroni correction. While this single-center study represents the first evidence of AI improving pneumothorax management workflow by simultaneously alerting both radiologists and referring clinicians, the system showed no significant improvement in time to tube thoracostomy (14.4 min; 95% CI, -35.0 to 63.9) or surgical consultation (86.3 min; 95% CI, -175.1 to 347.6), highlighting the need for multicenter validation. 

Care to Explain? AI Explanation Types Differentially Impact Chest Radiograph Diagnostic Performance and Physician Trust in AI Open Access

Radiology, published 11/19/24 | Estimated Read Time: 25 minutes

Study of 220 physicians demonstrates that local (feature-based) AI explanations significantly improve chest radiograph diagnostic accuracy compared to global (prototype-based) explanations when AI advice was correct (92.8% vs 85.3%, p<.001), while reducing time spent reviewing AI suggestions by 19% (p=.01). The impact varied by expertise level: non-radiologists showed enhanced performance with local explanations only when AI confidence was high (β=1.62, p=.002), while radiologists benefited from local explanations only with low AI confidence (β=1.12, p=.03), though the study's web-based format and limited cases with incorrect AI advice (25%) suggest the need for further clinical validation.

Application of deep learning techniques for breath-hold, high-precision T2-weighted magnetic resonance imaging of the abdomen

Abdominal Radiology, published 11/13/24 | Estimated Read Time: 15 minutes

Deep learning-based Precise IQ Engine (PIQE) reconstruction dramatically enhances T2-weighted abdominal MRI quality, with single-shot fast spin-echo (SS-FSE) sequences showing significantly higher liver SNR (11.8 vs 7.8, p<0.001) and improved contrast-to-noise ratio for nonsolid hepatic lesions (43.2 vs 34.5, p<0.001) compared to multi-shot FSE. While both PIQE-enhanced sequences demonstrated better image quality than standard protocols, the SS-FSE with PIQE emerged as superior for visualizing pancreatic ducts and cystic lesions, though the study's single-center design and lack of pathological confirmation warrant further investigation.

Deep Learning Algorithms for Breast Cancer Detection in a UK Screening Cohort: As Stand-alone Readers and Combined with Human Readers Open Access

Radiology, published 11/19/24 | Estimated Read Time: 20 minutes

Large UK study evaluating three commercial AI algorithms for mammography screening (n=26,722 cases) demonstrated that two algorithms achieved noninferior sensitivity compared to single human readers (DL-1: 64.8%, DL-3: 58.9% vs human: 62.8%; p<0.001), with one achieving superior specificity (DL-3: 97.9% vs 96.5%; p<0.001). When combined with human readers, all three AI systems matched double-reading accuracy while reducing recall rates, though they increased arbitration rates (5.2-9.5% vs 2.7%; p<0.001), suggesting AI could effectively supplement breast screening programs while maintaining reading standards across different sites and mammography vendors—though the retrospective nature and potential training data overlap remain important limitations.

Minimally interactive segmentation of soft-tissue tumors on CT and MRI using deep learning Open Access

European Radiology, published 11/19/24 | Estimated Read Time: 15 minutes

Deep learning-based method for soft-tissue tumor segmentation requires just six clicked points from users and achieves high accuracy (DSC: CT 0.85±0.11, MRI 0.84±0.12), while reducing segmentation time from ~30 minutes to 4-6 minutes compared to manual methods. The approach showed excellent generalization across unseen tumor types and imaging sequences, though validation was limited to only 14 of 100+ STT phenotypes and single imaging sequences rather than multimodal inputs.

Upcoming Conferences and Approaching Abstract Submission Deadlines

If you have any comments, questions, or feedback, please respond to this email!

Thank you,

Emily Barnard