BostonScience, Tech & Medicine

AI Breakthrough in Boston Unveils X-Ray Vision Capable of Detecting Lung Cancer in Non-Smokers

 AI Breakthrough in Boston Unveils X-Ray Vision Capable of Detecting Lung Cancer in Non-SmokersSource: Google Street View
Sam Cavanaugh
Published on November 24, 2023

It's a breathing breakthrough for non-smokers—a game-changing artificial intelligence (AI) tool has been developed to identify those who have never lit up but might still face the life-threatening risk of lung cancer, Boston scientists announced. The American Cancer Society has projected a staggering 238,340 new cases of lung cancer in the U.S. this year, with up to 20% of these affecting non-smokers according to a news report of PR Newswire.

According to a study spearheaded by savvy medical student Anika S. Walia from Boston University, presented at the Radiological Society of North America's (RSNA) annual meeting, "A major advantage to our approach is that it only requires a single chest-X-ray image", which is commonplace in medical tests and easily accessible in medical records, her innovation has the potential to overturn current Medicare and USPSTF guidelines that only recommend lung cancer screenings for those with a significant history of smoking—leaving non-smokers in a vulnerable blind spot as their cancer is frequently caught at an advanced stage yet AI could be the beacon of hope they need for early detection as reported by the MassLive.

The groundbreaking "CXR-Lung-Risk" model was trained with a mammoth dataset of 147,497 chest X-rays and has showcased promising results in identifying those at high risk even from a pool of never-smokers; these subjects’ records were trawled for precursor patterns to disease, sparking a potential revolution in lung cancer screening methods that could save countless lives by catching the illness sooner.

Senior author and director of artificial intelligence at Massachusetts General Hospital Michael T. Lu, M.D., M.P.H., highlighted, "Since cigarette smoking rates are declining, approaches to detect lung cancer early in those who do not smoke are going to be increasingly important" his vision, backed by the study’s findings where 2.9% of the high-risk individuals were later diagnosed with the cancer underlines the sheer urgency and potential impact of this novel tech, especially when you consider the sharp decline in smoking rates from 20.9% in 2005 to just 11.5% in 2021 among U.S. adults as these lung cancer cases rise conspicuously amongst non-smokers based on the alarming facts stated in MassLive's report.

The researchers' mission, infused with support from various academic bodies, including the National Academy of Medicine, paves the way for an 'opportunistic screening' of lung cancer, capitalizing on existing chest X-rays to flag high risks among the most unlikely patients, which may well redefine the landscape of cancer screening and early intervention for the better.

BostonScience, Tech & Medicine