Study: Liver, spleen and coronary artery analysis used to predict severity.

Helen Hwang, Contributor

December 5, 2022

2 Min Read
COVID-19 coronavirus
Getty Images

A new study has shown that deep learning can be used to determine the severity of COVID-19 by analyzing a patient’s liver fat.

Dubbed ‘DeHFt,' the deep learning-based system analyzed measurements of liver fat from patients with nonalcoholic fatty liver disease, called hepatic steatosis. The scientists linked the data to COVID and found that the patients were one and a half times more likely to experience severe COVID.

“We know that hepatic steatosis is a risk factor for COVID-19. Now we can use this pipeline to identify high-risk patients and based upon that, clinicians can make better-informed decisions about levels of care and the early use of therapeutics, such as antivirals,” said Gourav Modanwal, the study’s first author and researcher in the Wallace H. Coulter Department of Biomedical Engineering at Emory University School of Medicine and Georgia Institute of Technology College of Engineering.

The machine learning platform used data from coronary artery calcium CT scans, which measure calcium-rich plaque in a person’s arteries. The images from the patient’s heart, liver and spleen were analyzed for hepatic fat, a process that’s challenging to measure because radiologists need higher magnification and resolution for proper assessments.

The DeHFT process also incorporates a two-step analysis. The first phase requires a segmentation model of the liver and spleen to be trained, based on the coronary artery calcium CT scans. A CT attenuation, which looks at the intensity of the liver and spleen, models 3D slices of the liver and spleen. When a liver presents as low intensity, it shows more fat infiltration. The spleen is the control image.

DeHFT was found to evaluate liver fat levels with better accuracy than a team of radiologists who showed variability in their scan assessments.

The study was published in The Lancet’s journal, eBioMedicine. The research team from Emory University and University Hospitals Cleveland collaborated with sites at Case Western Reserve University, Wuhan University, the Atlanta Veterans Administration Medical Center and Guangdong Academy of Medical Sciences.

“Our study suggests that machine learning based on routine CT scans can help in accurate quantification of liver fat, which has implications that extend beyond COVID-19 severity assessment,” said Anant Madabhushi, the study’s senior author and professor in the Wallace H. Coulter Department of Biomedical Engineering at Emory School of Medicine and Georgia Institute of Technology College of Engineering.

About the Author(s)

Helen Hwang

Contributor, AI Business

Helen Hwang is an award-winning journalist, author, and mechanical engineer. She writes about technology, health care, travel, and food. She's based in California.

Keep up with the ever-evolving AI landscape
Unlock exclusive AI content by subscribing to our newsletter!!

You May Also Like