Radiology in public focus

3D MRI Technique Guides Precision Treatment of Kids’ Heart Condition

Researchers at Children’s Hospital of Philadelphia (CHOP) developed 3D volume rendering methods for cardiac MRI that show how blood moves through complex heart structures, including valve leaflets. The MRI technique may aid in treatment decisions in children with congenital heart defects, according to new research published in Radiology: Cardiothoracic Imaging.

“In patients with holes in the heart structure or leaflets that don’t form a complete seal, we can now see the valve leaflets moving and identify exactly where a valve is leaking, which has not been possible with MRI before this technique,” said study coauthor Matthew Jolley, MD, a pediatric cardiac anesthesiologist and cardiologist at CHOP and an associate professor at the University of Pennsylvania in Philadelphia.

The approach may be especially valuable for pediatric patients, who often require detailed imaging to plan complex repairs early in life.

Read the related RSNA News story.

Deepfake X-Rays Fool Radiologists and AI

Neither radiologists nor multimodal large language models (LLMs) can easily distinguish AI-generated “deepfake” X-ray images from authentic ones, according to a study published in Radiology.

Led by Mickael Tordjman, MD, post-doctoral fellow, Icahn School of Medicine at Mount Sinai, NY, the retrospective study included 17 radiologists from six countries with experience ranging from 0 to 40 years. Participants reviewed 264 X-ray images, half authentic and half AI-generated using ChatGPT and Roentgen diffusion model.

When unaware that synthetic images were included, only 41% of participants spontaneously identified AI-generated images. After being informed, mean accuracy rose to 75%. The performance of four multimodal LLMs ranged from 57% to 85%, with even the image-generating model unable to detect all deepfakes.

“We are potentially only seeing the tip of the iceberg,” Dr. Tordjman said. “The logical next step in this evolution is AI-generation of synthetic 3D images, such as CT and MRI. Establishing educational datasets and detection tools now is critical.”

Read the related RSNA News story.

Photon-counting CT Outperforms Conventional CT in Lung Cancer Management

Photon-counting CT (PCCT) may reduce radiation exposure and adverse reactions while improving image quality and cancer detection in lung imaging, according to a study published in Radiology.

In the prospective study, researchers compared the benefits of contrast-enhanced chest CT using ultra-high-resolution PCCT versus conventional CT on a cohort of 200 individuals with lung cancer. Compared to conventional CT, PCCT reduced radiation and iodine exposure by 66.34% and 26.57%, respectively.

“We believe photon-counting CT might replace conventional CT in the near future due to its improved imaging quality and the diagnostic confidence it offers,” said study author Songwei Yue, MD, chief physician, professor and deputy director of radiology in the Department of Radiology at The First Affiliated Hospital of Zhengzhou University in Henan, China.

Read the related RSNA News story.