Get all your news in one place.
100’s of premium titles.
One app.
Start reading
Zenger
Zenger
Health
James Gamble

AI Identifies Non-Smokers At Risk Of Lung Cancer

AI technology could play a crucial role in identifying non-smokers at risk of developing lung cancer. ANNA SHVETS/PEXELS

AI technology could play a crucial role in identifying non-smokers at risk of developing lung cancer.

Scientists in the United States used an artificial intelligence (AI) model to identify non-smokers at a higher risk of developing lung cancer by viewing a single X-ray of their chests.

They found that the android model accurately identified a group of around 30 percent out of nearly 17,500 patients who had a 2.1 times higher risk of developing lung cancer.

AI technology could play a crucial role in identifying non-smokers at risk of developing lung cancer. ANNA SHVETS/PEXELS

The research team hailed the new project as being able to identify at-risk patients of what is the world’s deadliest cancer, as it is currently difficult to predict this risk in those who do not or have rarely smoked.

The study, which will be presented at the annual meeting of the Radiological Society of North America (RSNA) in Chicago, sought to find a way of identifying people at risk of developing lung cancer, despite abstaining from smoking.

Lung cancer is the most common cause of cancer death in the world, accounting for nearly half of all new cancer cases in the United States along with breast, bronchus, prostate, and colorectal cancers.

The American Cancer Society estimates that around 238,340 new cases of lung cancer will occur this year alone, as well as around 127,000 lung cancer deaths.

Approximately 10 to 20 percent of lung cancers occur in those who have either never smoked cigarettes or smoked less than a hundred cigarettes in their life.

However, despite the increasing prevalence of lung cancer occurrence in these groups, they are often not recommended to get screened, with smokers instead prioritized.

The United States Preventive Services Task Force (USPSTF) currently only recommends lung cancer screening for adults aged 50 to 80 who have at least a 20-pack-year smoking history – having smoked a pack of cigarettes a day for the past 20 years, or two packs a day for the past ten years – and currently smoke or have quit within the past 15 years.

The USPSTF does not recommend screening for individuals who have never smoked or who have smoked very little.

The study’s lead author, Anika Walia, a medical student at Boston University School of Medicine and researcher at the Cardiovascular Imaging Research Center (CIRC) at Massachusetts General Hospital (MGH) and Harvard Medical School in Boston, explained that due to the reduced screening, when lung cancer is discovered in non-smokers, it’s often more advanced than those found in smokers.

“Current Medicare and USPSTF guidelines recommend lung cancer screening CT only for individuals with a substantial smoking history,” she said.

“However, lung cancer is increasingly common in never-smokers and often presents at an advanced stage.”

AI technology could play a crucial role in identifying non-smokers at risk of developing lung cancer. LIL ARTSY/PEXELS

One reason non-smokers are excluded from screening recommendations is that it’s difficult to predict lung cancer risk in them.

Existing lung cancer risk scores require information that is not readily available for most individuals, such as family history of lung cancer, pulmonary function testing or serum biomarkers.

Therefore, the CIRC researchers set out to improve lung cancer risk prediction in non-smokers by using a ‘deep learning’ AI model, which could identify those most at risk by observing chest X-rays of patients.

Deep learning is an advanced type of AI that can be trained to search X-ray images to find patterns associated with disease.

“A major advantage to our approach,” Walia said, “is that it only requires a single chest-X-ray image – which is one of the most common tests in medicine and widely available in the electronic medical record.”

The ‘CXR-Lung-Risk’ model was developed using nearly 150,000 chest X-rays of 40,643 asymptomatic smokers and never-smokers from the Prostate, Lung, Colorectal, and Ovarian (PLCO) cancer screening trial to predict lung-related mortality risk, based on a single chest X-ray image.

The model’s results were validated in a separate group of never-smokers having routine outpatient chest X-rays between 2013 and 2014.

The primary outcome was a six-year incident of lung cancer, identified using the International Classification of Disease codes.

Risk scores were then converted to low, moderate and high-risk groups based on externally derived risk thresholds.

Out of 17,407 patients with a mean age of 63, 28 percent were deemed high-risk by the deep learning model, and 2.9 percent of these patients later had a diagnosis of lung cancer.

The high-risk group well exceeded the 1.3 percent six-year risk threshold where lung cancer screening is recommended by National Comprehensive Cancer Network guidelines.

After adjusting for age, sex, race, previous lower respiratory tract infection and prevalent chronic obstructive pulmonary disease, there was still a 2.1 times greater risk of developing lung cancer in the high-risk compared to the low-risk group.

“This AI tool opens the door for opportunistic screening for never-smokers at high risk of lung cancer, using existing chest X-rays in the electronic medical record,” senior author Dr. Michael Lu said.

The director of artificial intelligence and co-director of CIRC at MGH added: “Since cigarette smoking rates are declining, approaches to detect lung cancer early in those who do not smoke are going to be increasingly important.”

Produced in association with SWNS Talker

Sign up to read this article
Read news from 100’s of titles, curated specifically for you.
Already a member? Sign in here
Related Stories
Top stories on inkl right now
One subscription that gives you access to news from hundreds of sites
Already a member? Sign in here
Our Picks
Fourteen days free
Download the app
One app. One membership.
100+ trusted global sources.