AI could help to diagnose lung cancer earlier

Lung cancer is the most common cause of cancer death, with around 1.8 million lives lost around the world each year. It is often diagnosed at a later stage when treatment is less likely to succeed.

But researchers worldwide hope that using AI to support lung cancer screening could make the process quicker and more efficient, and ultimately help diagnose more patients at an early stage.

Computerised tomography, or CT scans, are already used to spot signs of lung tumours, followed by a biopsy or surgery to confirm whether the tumour is malignant. However, each scan involves an expert radiologist examining around 300 images and looking for signs of cancer that can be small.

Trials using CT scans to screen people with a high risk of lung cancer have shown promise, but screening is hindered by the practical difficulty of a radiologist reviewing each image, one at a time, to determine who needs further tests.

Researcher Benoît Audelan said: “Screening for lung cancer would mean many more CT scans being taken and we do not have enough radiologists to review them all. That’s why we need to develop computer programs that can help. Our study shows that this program can find potential signs of lung cancer up to a year earlier.

“The aim of our research is not to replace radiologists but to assist them by giving them a tool that can spot the earliest signs of lung cancer,” added Audelan, who is part of the Epione project team of the Inria (France’s National Institute for Research in Digital Science and Technology) centre at Côte d’Azur University.

In the new study, led by Audelan, the research team trained their AI program using a set of CT scans from 888 patients that had already been examined by radiologists to identify suspicious growths.

They then tested it on a fresh set of 1,179 patients who were part of a lung screening trial with a three-year follow-up, using CT scans that were taken in the last two years of the trial. These included 177 patients who were diagnosed with lung cancer via a biopsy after their final scan in the trial.

As a result, the program identified 172 of the 177 malignant tumours in those CT scans – it was 97 per cent effective in detecting cancers. The five tumours that the program missed were near the centre of the chest, where tumours are more challenging to distinguish from healthy parts of the body.

The researchers also tested the program on scans taken a year before the tumours were diagnosed in the same 1,179 patients. Here, it could identify 152 suspicious areas that were later diagnosed as cancer.

But the team warned the program also identifies too many suspicious areas that are not cancer (false positives) and this would need to be vastly improved before the program could be used in the clinic because investigating all of these would cause unnecessary biopsies.

The researchers plan to work on a new system that will be better able to differentiate between malignant and non-malignant tissue to help radiologists decide which patients should be investigated further.