Advertisement

AI could help doctors detect lung cancer at earlier stage

A new AI software could help to doctors to detect lung cancer at an earlier stage (file picture) (PA Archive)
A new AI software could help to doctors to detect lung cancer at an earlier stage (file picture) (PA Archive)

A new AI software could help to doctors to detect lung cancer at an earlier stage and deliver more effective treatments for patients.

The DOLCE study, supported by the National Institute for Health and Care Research, is investigating a new software programme to help analyse whether spots on the lung found in CT scans are cancerous.

Researchers hope the technology could help to save lives and help the NHS save money by avoiding unnecessary procedures.

Lung cancer is the deadliest type of cancer. In the UK, 47,000 people are newly diagnosed each year and the disease causes over 35,000 deaths. Catching the disease early is key to preventing serious illness and death.

Dr Richard Lee, a Consultant Physician in Respiratory Medicine at The Royal Marsden and the NIHR’s national specialty lead for screening, prevention and diagnosis, said: “This technology could help us to spot cancer at an earlier stage, when treatment is likely to be more effective, by allowing us to invite people for investigation sooner.

“Often, when we find spots on the lung from a CT scan, the person will require follow-up imaging, which can take many months and can cause worry. The hope is an earlier AI decision could help to speed things up.

“A current challenge for consultants is correctly identifying lung cancer from these spots on the CT scans. An automated approach could reduce variability and provide a safety net for these decisions. The software offers us many opportunities, but research is critical to ensure that we make the most of these evolving technologies with the scientific proof that they are effective.”

The study is led by Chief Investigator Professor David Baldwin at Nottingham University Hospitals in partnership with Optellum lung AI software.