Radiological Society of North America (RSNA)

In patients with congestive heart failure, pulmonary edema is a leading cause of hospitalization. Imaging is critical for assessing the severity of this condition and determining the optimal treatment and monitoring plan. In an article published in Radiology Advances, Viacheslav V. Danilov, PhD, Pompeu Fabra University, Barcelona, Spain, and colleagues sought to develop a deep learning method for identifying radiographic features associated with pulmonary edema. For this retrospective study, the researchers used a dataset of chest radiograph images from 741 patients suspected of pulmonary edema. An experienced radiologist annotated various features of edema including cephalization, Kerley lines, pleural effusion, bat wings and infiltrate features of edema. The methodology involved two stages: lung segmentation using an ensemble of three object detection networks, and edema feature localization which evaluated eight object detection networks, assessing their performance with average precision (AP) and mean AP. Effusion, infiltrate, and bat wing features were best detected by the Side-Aware Boundary Localization (SABL) network with corresponding APs of 0.599, 0.395, and 0.926, respectively. Furthermore, SABL achieved the highest overall mean AP of 0.568. The Cascade Region Proposal Network attained the highest AP of 0.417 for Kerley lines and the Probabilistic Anchor Assignment network achieved the highest AP of 0.533 for cephalization. “The proposed methodology, with the application of SABL, Cascade Region Proposal Network, and Probabilistic Anchor Assignment detection networks, is accurate and efficient in localizing and identifying pulmonary edema features and is, therefore, a promising diagnostic candidate for interpretable severity assessment of pulmonary edema,” the authors conclude. Read the complete study, “Explainable AI to identify radiographic features of pulmonary edema.” https://bit.ly/3Y04X0z

Initiatives to create and maintain sustainable radiological practices can start small and then gain momentum. https://bit.ly/4f6vsY7

Registration is now open for RSNA 2024! Join us in Chicago, Dec. 1-5, where we are Building Intelligent Connections and exploring how to intentionally and intelligently build the right connections between people and technology. Register today: RSNA.org/Annual-Meeting

Special report addresses key factors in the successful application of AI in clinical radiology practice. https://bit.ly/3XYU52Y

Learn alongside expert faculty in Santiago, Chile, to enhance your skills in four critical areas—neuroradiology, chest imaging, acute abdominal imaging and musculoskeletal imaging at The Current State of Emergency Radiology: An Interactive Course. https://bit.ly/4di9O2c

No matter where you are in your AI education, RSNA can help take you to the next step through the high quality, proven course structures in the Foundational, Advanced and Emergency Certificate courses. https://lnkd.in/eS3vcRV3

Cesarean section rates globally have increased, reaching 50% in some countries. The risk of abnormal implantation from prior cesarean scars poses severe complications, including maternal and fetal mortality and fertility loss. Subsequent cesarean scar ectopic pregnancies (CSEPs) are on the rise, particularly in recent decades. Underdiagnosis and underreporting make the true incidence unclear, leading to high maternal morbidity and mortality rates. In an article published in RadioGraphics, Anne Kennedy, MBBCh, BAO, and Paula Woodward, MD, University of Utah, Salt Lake City, and colleagues review the signs of CSEP at imaging, some pitfalls that may lead to delayed or missed diagnosis, and the resulting consequences. The authors emphasize the importance of differentiating CSEPs from low implantation of a normal pregnancy, cervical ectopic pregnancy and evolving pregnancy loss. Early recognition of CSEPs allows for prompt and safe treatment that is usually surgical and can reduce health care costs and hospital stays and preserve fertility. Early diagnosis and treatment of CSEP can also avert the need for a hysterectomy, lessening psychological impacts. “The most important thing that radiologists can do to avoid delayed or missed diagnosis of CSEP is to routinely check for a history of cesarean section. If there are findings of concern for CSEP, prompt referral for further evaluation and treatment is mandatory. Treatment early in the first trimester is safe and effective and allows preservation of fertility in women who desire future pregnancy. Management of CSEP is determined on an individual basis in all cases,” the authors conclude. Read the complete article, “Cesarean Scar Ectopic Pregnancy: A Do-Not-Miss Diagnosis.” This article is also available for CME at RSNA.org/Learning-Center. Follow the RadioGraphics editor on X @RadG_Editor. https://bit.ly/4fbhdl0

Get hands-on experience in the entire grant preparation process through the Advanced Course in Grant Writing. Apply by July 22! https://bit.ly/3JBEFrm

Cutting-Edge Research call for abstracts is officially open and will close on Aug. 7, noon CT. Do not miss your opportunity to share your innovative research at RSNA 2024! https://bit.ly/3HviHHd

Happy Birthday, Nikola Tesla, born this day in 1856! Learn more about Tesla's contributions to radiology: http://bit.ly/1SQc4yh