Research finds AI can identify lung cancer years before symptoms show

The findings presented at the 2024 World Conference on Lung Cancer found that Qure.ai’s qXR system can identify cancerous nodules years before symptoms arrive.

Joshua Silverwood September 11 2024

Research has found that artificial intelligence (AI) chest X-rays can detect the early symptoms of lung cancer up to three years before symptoms present themselves.

The study, presented at the International Association for the Study of Lung Cancer (IASLC) 2024 World Conference on Lung Cancer in San Diego, found that AI-powered chest X-ray interpretation can detect pulmonary nodules which could develop into early-stage lung cancers before symptoms present themselves.

The retrospective study demonstrated via interim results, an average diagnostic delay of nearly three years from the first abnormal chest X-ray. The study, conducted at Bangkok’s Phrapokklao Hospital’s Cancer Centre of Excellence, used the Qure.ai algorithm, qXR, to detect the cancerous nodules.

Passakorn Wanchaijiraboon, investigator for the study, said; "This abstract study, presented at the World Conference on Lung Cancer, provides a snapshot of the significant potential that AI-assisted chest X-ray analysis holds for transforming early cancer detection and reducing the rate of missed lung cancer diagnoses.

“In most Thai government hospitals, chest X-rays are interpreted by non-radiologists. However, in community hospitals, there are often no radiologists available to read chest X-rays at all. By overlaying specialist AI to read all cases, we can support clinicians in detecting incidental high-risk nodules that may lead to lung cancer.

“This approach can streamline decision-making and potentially improve patient survival through the earlier diagnosis of cancer.”

Research published in 2017 found that Lung Cancer has one of the poorest survival outcomes of all cancers, with over two-thirds of patients diagnosed at an advanced stage.

Researchers retrospectively evaluated the chest X-ray image database of newly diagnosed lung cancer patients over an annual period using qXR. Missed lung cancer was defined as missed in the original report six months prior to a definitive lung cancer diagnosis.

They found that 18% of patient cases were found to have a missed lung cancer diagnosis over an average period of nearly three years. Half of all cases had chest X-rays taken for non-respiratory symptoms as part of a health check-up.

Bhargava Reddy, chief business officer for oncology at Qure.ai, said: “This is an exciting evidence example that underscores the transformative potential of AI in the fight against lung cancer. Overlaying AI on chest X-rays casts the net wider by proactively triaging patients for the risk of lung cancer. It goes beyond people with symptoms or qualifying for screening initiatives based on age or smoking history, to currently invisible and unprofiled patient populations thus detecting lung cancers earlier.”

Last month Qure.AI was able to secure US Food and Drug Administration (FDA) clearance for its computed tomography (CT) imaging solution qCT LN Quant. Meanwhile, US-based competitor Invenio Imaging revealed it had completed enrolment for a pivotal trial of its AI lung cancer identification software.

Uncover your next opportunity with expert reports

Steer your business strategy with key data and insights from our latest market research reports and company profiles. Not ready to buy? Start small by downloading a sample report first.

Newsletters by sectors

close

Sign up to the newsletter: In Brief

Visit our Privacy Policy for more information about our services, how we may use, process and share your personal data, including information of your rights in respect of your personal data and how you can unsubscribe from future marketing communications. Our services are intended for corporate subscribers and you warrant that the email address submitted is your corporate email address.

Thank you for subscribing

View all newsletters from across the GlobalData Media network.

close