A new study is to investigate how artificial intelligence (AI) could be used to improve early diagnosis of pancreatic cancer.

Using data from patients who have been diagnosed with the disease, researchers from a group of universities including Manchester Metropolitan will use AI to identify potential biomarkers and risk groups who are more likely to develop pancreatic cancer in the future.

If successful, clinicians will be able to detect early signs of the disease before symptoms start to appear, at which point the cancer is usually too advanced for treatment.

The study, led by Edge Hill University, has been awarded a GBP100,000 start-up grant from Cancer Research UK, Pancreatic Cancer UK and the Engineering and Physical Sciences Research Council (EPSRC), part of UK Research and Innovation (UKRI).

Reader in Computer Vision, Dr Moi Hoon Yap and her team at Manchester Met will be working to develop a method to explain the computer model to clinicians - ensuring they can interpret the data and clearly understand why the system has detected certain individuals as being at risk of the disease.

She said: 'For AI to be used as a tool to help the diagnosis of pancreatic cancer, healthcare professionals must be able to trust it. This means the data generated by the AI must be transparent, with clinicians not only able to fully understand the results and why they have been generated, but they must also agree with this data.

'Our aim is to find a way, using a method called Explainable AI, to ensure the data produced by our computer model can be explained and is reliable to clinicians, fully supporting them in making their own diagnosis of a patient.'

According to Cancer Research UK, there are around 10,300 new pancreatic cancer cases in the UK every year, with around eight in 10 cases being diagnosed at a late stage in England and Scotland.

Polling for Pancreatic Cancer UK revealed that just one in 10 GPs say they have the tools they need to diagnose pancreatic cancer early enough for treatment to be possible.

Dr Ardhendu Behera is a Reader in Computer Vision & AI at Edge Hill University, said: 'Pancreatic cancer progresses very rapidly and because no method of early detection has been discovered, it is one of the most dangerous types of cancer. Symptoms are usually very vague and do not appear until the cancer is in its later stages, at which point it is sadly too late to treat effectively in most cases.

'This is a data-driven approach that aims to find out whether it is possible to use AI to signal that there are links between risk groups and those who have already been diagnosed. Using routinely collected data such as CT/MRI, health conditions, pathology, and blood tests, the AI can identify the possible predictors of pancreatic cancer and will screen out people at high-risk. The potential of AI is incredible, and in this instance could help to save and extend the lives of patients.'

The year-long study will also involve Queen Mary University of London, as well as researchers from University College London and the University of Hertfordshire.

(C) 2021 Electronic News Publishing, source ENP Newswire