AI improves image quality for diagnosis of liver lesions
AI can improve CT findings and play a key role in the evaluation of hypovascular hepatic metastases, according to a new study published in Radiology: Artificial Intelligence.
“CT studies that evaluate the liver, abdomen and chest simultaneously in a single session are the primary modality for conducting follow-up examinations and for determining the disease stage,” wrote Yuko Nakamura, department of diagnostic radiology at Hiroshima University in Japan, and colleagues. “However, as the detectability of hepatic metastases, especially those smaller than 10 mm, on contrast material—enhanced CT images remains limited, further improvements are needed.”
Nakamura et al. noted this is where image reconstruction algorithms enter the equation. Model-based iterative reconstruction (MBIR), for instance, is an advanced algorithm specifically for CT studies that has been found to improve image quality. MBIR, however, has been untested in certain situations and hybrid iterative reconstruction (IR) is often used as an alternative, even if it is as viewed as “inferior.”
Could deep learning-based reconstruction (DLR) be used in place of MBIR or IR when treating patients with hypovascular hepatic metastases? To explore that very question, the authors analyzed data from 58 patients with hypovascular hepatic metastases. The smallest liver lesion in each patient was evaluated, and the authors compared the performance of hybrid IR images and DLR images.
Overall, the DLR images had “significantly lower” image noise and a “significantly higher” contrast-to-noise ratio (CNR). Processing DLR images was also roughly three to five times faster than other reconstruction methods.
“DLR resulted in quantitative and qualitative improvements of abdominal CT images acquired for the evaluation of hypovascular hepatic metastases,” the authors wrote. “Future studies will focus on the application of DLR to different clinical scenarios, to different patient populations with a diverse body habitus, and to different acquisition techniques involving low radiation doses and high spatial resolution techniques.”
The research did have certain limitations, the team added. The study population was “relatively small,” for example, and data from only one institution was evaluated. More research is also needed to see if DLR can help with the diagnosis of “other hepatic tumors such as hepatocellular carcinoma.”
Michael has more than 18 years of experience as a professional writer and editor. He has written at length about cardiology, radiology, artificial intelligence and other key healthcare topics.