A team of researchers from University of Glasgow have worked alongside colleagues in New York to develop an AI system to help improve cancer diagnosis.

The new system, called ‘Histomorphological Phenotype Learning’ (HPL), has been spotting the signs of the disease in tests with ‘remarkable accuracy’, according to those involved.

An international team of AI specialists and cancer scientists are behind the breakthrough and it’s believed it can also help provide reliable predictions of patient outcomes.

Currently, pathologists examine tissue samples under a microscope to provide doctors with support on how to treat each patient and help them with an idea of their recovery chances.

But the AI could make that much quicker for everyone involved and help pathologists get results to doctors and therefore onto patients in much quicker ways.

The team have outlined how they have developed and trained the system in a new paper released in the Nature Communications journal.

They started their research by collecting high resolution images of tissue samples from 452 patients stored in the United States National Cancer Institute’s database and then developed an algorithm to analyse the images and spot patterns based solely on the visual data in each slide.

The papers senior author was Dr Ke Yuan and said: “We didn’t provide the algorithm with any insight into what the samples were or what we expected it to find. Nonetheless, it learned to spot recurring visual elements in the tiles which correspond to textures, cell properties and tissue architectures called phenotypes.

“By comparing those visual elements across the whole series of images it examined, it recognised phenotypes which often appeared together, independently picking out the architectural patterns that human pathologists had already identified in the samples.”

When the team then added analysis of slides from squamous cell lung cancer it was able to tell the difference between that and the lung adenocarcinoma which was in the initial data to 99% accuracy.

The predictions made by the system linked well with the outcome of real life patients and correctly assessed the likelihood and timing of any cancer return to 72%, which is better than human pathologists at 64%.

The research was then broadened to included other types of cancer including breast, prostate and bladder and returned similar results as the initial tests.

Professor John Le Quesne said: We were surprised but very pleased by the effectiveness of machine learning to tackle this task. It takes many years to train human pathologists to identify the cancer subtypes they examine under the microscope and draw conclusions about the most likely outcomes for patients. It’s a difficult, time-consuming job, and even highly-trained experts can sometimes draw different conclusions from the same slide.

“In a sense, the algorithm at the heart of the HPL system taught itself from first principles to speak the language of cancer – to recognise the extremely complex patterns in the slides and ‘read’ what they can tell us about both the type of cancer and its potential effect on patients’ long-term health. Unlike a human pathologist, it doesn’t understand what it’s looking at, but it can still draw strikingly accurate conclusions based on mathematical analysis.

“It could prove to be an invaluable tool to aid pathologists in the future, augmenting their existing skills with an entirely unbiased second opinion. The insight provided by human expertise and AI analysis working together could provide faster, more accurate cancer diagnoses and evaluations of patients’ likely outcomes. That, in turn, could help improve monitoring and better-tailored care across each patients’ treatment.”