Radiology in public focus

High Accuracy AI Improves Lung Cancer Detection

Assistance from an AI algorithm with high diagnostic accuracy improved radiologist performance in detecting lung cancers on chest X-rays and increased human acceptance of AI suggestions, according to a study published in Radiology.

In the retrospective study, led by Chang Min Park, MD, PhD, Department of Radiology and Institute of Radiation Medicine at Seoul National University College of Medicine, use of the high accuracy AI improved readers’ detection performance to a greater extent than low-accuracy AI and led to more frequent changes in reader determinations.

Read the related RSNA News story, “High Accuracy AI Improves Lung Cancer Detection.”

Photon-Counting CT Can Evaluate Lung Function

New CT technology allows for a comprehensive, simultaneous evaluation of lung structure and function, something not possible with standard CT, according to a study published in Radiology.

Researchers in Germany and the Netherlands developed a chest imaging protocol using photon-counting CT technology that yields information on structure and function of the lungs as an all-in-one procedure.

According to study senior author Hoen-oh Shin, MD, professor of radiology at the Institute of Diagnostic and Interventional Radiology at Hannover Medical School in Germany, and colleagues, the protocol enables high image quality at a radiation dose below that of a standard chest CT. It also provides better spatial resolution and options for spectral imaging and requires advanced software but no additional hardware.

Read the related RSNA News story, “Photon-Counting CT Can Evaluate Lung Function.”

AI Uses Lung CT Data to Predict Mortality Risk

AI can use data from low-dose CT (LDCT) scans of the lungs to improve risk prediction for death from lung cancer, cardiovascular disease and other causes, according to a study published in Radiology.

In a prior study, lead author Kaiwen Xu, a PhD candidate in the Department of Computer Science at Vanderbilt University in Nashville, TN, and colleagues developed, tested and publicly released an AI algorithm that automatically derives body composition measurements from lung screening LDCT.

For the new study, the researchers assessed the added value of the AI-derived body composition measurements.

Results showed that including these measurements improved risk prediction for death from lung cancer, cardiovascular disease and all-cause mortality. Measurements associated with fat found within a muscle were particularly strong predictors of mortality.

Read the related RSNA News story, “AI Uses Lung CT Data to Predict Mortality Risk.”