AI scanner could reduce lung cancer diagnosis time to ‘minutes’

Scottish researchers have developed an artificial intelligence (AI) powered method for accurately predicting lung cancer without the need for slower, more expensive lab techniques. 

The technology, developed by scientists at the University of Edinburgh and NHS Lothian, uses a technique called fluorescence lifetime imaging microscopy (FLIM) to predict specific DNA genetic changes, such as mutations in the EGFR gene, which can be a key indicator of lung cancer before it takes hold.  

The FLIM technique captures natural light signals from tissue samples, which are then analysed by artificial intelligence for patterns that can predict cancer. In the study, the method was able to predict the presence of EGFR mutations and also distinguish between the two most common types of EGFR mutations that clinicians use to make treatment decisions. 

“This approach has the potential to take processes that currently cost thousands of pounds and require weeks of lab work and reduce them to something that takes minutes and costs hundreds,” said Dr Qiang Wang, co-lead of the study from the Institute for Regeneration and Repair.

“That is a step change in what is clinically achievable, particularly for centres and health systems where access to complex molecular testing is limited.” 

According to the most recent data from Public Health Scotland (PHS), lung cancer remained the most common cause of cancer death overall in 2024, with over 3,650 deaths. This accounted for around a fifth of all cancer deaths in Scotland. According to PHS, most of these deaths could be avoided by eliminating smoking. 

The process for detecting the mutations that can indicate lung cancers currently requires laboratory tests like gene sequencing, which can be expensive, time-consuming, and use valuable tissue from small biopsy samples that are limited in number.  

By using the AI-powered approach, the researchers say the technology has the ability to speed up diagnosis while also preserving limited biopsy material due to the method not damaging the lung tissue, leaving it intact and available for further analysis.   

Dr David Dorward, consultant thoracic pathologist at NHS Lothian, said: “Clinicians are increasingly seeing more patients with earlier-stage disease and dealing a growing number of biopsy samples, placing significant pressure on diagnostic services. Technologies like this, which can deliver more information from smaller tissue samples at speed, will be essential for developing clinically effective diagnostic pathways.”   

A similar AI scanning tool named GEMINI delivered by NHS Grampian and the University of Aberdeen, analysed over 10,000 mammograms in 2023, with research finding that the technology increased detection rates of breast cancer by over 10 per cent and reduced the time to notify affected women from 14 days to just three. 

Earlier this year the innovation agency InnoScot Health highlighted that integrating artificial intelligence into the NHS represents a “significant challenge” for decision-makers, despite the “huge promise” that the technology offers, warning that there must be a balance between the opportunities and the need for good governance and building public trust.