AI Blood Testing Technology Could Change How We Detect Lung Cancer

Lung cancer is the most lethal type of cancer, with almost 2 million lung cancer patients dying every year around the world. While low-dose CT screening is estimated to potentially prevent tens of thousands of deaths, fewer than 6% of Americans at risk for lung cancer are screened due to concerns over radiation exposure, complications from invasive procedures, and the possible harm caused by investigating false positive results.

A novel AI (artificial intelligence) blood testing technology could change that. In a recently published study, th ~ new technology detected more than 90% of lung cancers in samples collected from roughly 800 individuals with and without cancer.

In a study published in Nature Communications, the study looked at an approach called DELFI (DNA evaluation of fragments for early interception), this method can pinpoint unique patterns in the fragmentation of the DNA shed from cancer cells circulating in the bloodstream.

Researchers tested DELFI in combination with clinical risk factors, a protein biomarker, and CT imaging using a cohort of 796 individuals. They found that this test helped detect 94% of patients with cancer at various stages and across a variety of SUbtypes. If this proves to be successful in other studies the authors believe this test could cut down on complicated and invasive approaches to diagnosis.