15-07-2025
Bowel cancer survivor backs research into new AI tool
John Keryakoplis, a 69-year-old grandfather from Flintshire, is backing a new Cardiff University study exploring how AI could identify which bowel cancer patients may not need life-altering surgery.
Mr Keryakoplis, has two grandchildren, was diagnosed with stage one bowel cancer in January 2020 after taking part in the NHS Wales bowel screening programme.
John Keryakoplis with his wife (Image: Supplied) He said: "It didn't feel great to be told I had cancer, but I was very reassured when the doctor said it was in its early stages and very treatable."
He underwent a week of radiotherapy in May 2020, followed by surgery in November 2021 to remove the tumour.
As part of his treatment, Mr Keryakoplis had a stoma fitted, which he named "George" after his father.
He said: "George the stoma even came with me to Menorca.
"I wore swimming shorts pulled up so high I looked like Simon Cowell."
Despite his diagnosis and treatment, he continued to swim throughout, and has now been cancer-free for five years.
He has also raised more than £1,000 for Cancer Research UK through swimming challenges.
Mr Keryakoplis said: "I try to live as healthily as I can after my diagnosis.
"This new research sounds really promising.
"It would be brilliant for bowel cancer patients to have a more accurate and personalised treatment plan, which could prevent people from having surgery if they don't need it."
Cardiff University's Professor Spezi has received a grant to develop an AI tool to improve the diagnosis and treatment of rectal cancer (Image: Matthew Horwood) The research he is supporting is led by Professor Emiliano Spezi at Cardiff University.
Professor Spezi has received a £500,000 Stand Up To Cancer grant to develop an AI tool to improve the diagnosis and treatment of rectal cancer, a type of bowel cancer.
Current treatment decisions are often based on invasive biopsies, but Professor Spezi's AI algorithm analyses 3D medical images to extract detailed data that could help doctors decide on the most effective treatment.
He said: "This research could lead to significantly improved and personalised treatment plans for patients in the future.
"We want to extract quantitative information from images that can't be seen with the naked eye.
"It could be transformative in the way rectal tumours are assessed and treatment is allocated.
"We should be able to identify a subgroup of patients for whom surgery is essential – and those for whom it may not be necessary – ultimately improving quality of life."
The technology can assess tumour volume and shape, distinguish between healthy and cancerous tissue, and identify complex patterns using deep learning.
Professor Spezi is collaborating with Professors Campbell Roxburgh and Joanne Edwards at the University of Glasgow.
The research is supported by Stand Up To Cancer, a joint fundraising campaign by Cancer Research UK and Channel 4, which aims to accelerate breakthroughs in cancer diagnosis and treatment.
Around 20,200 people in Wales are diagnosed with cancer each year.
Visit to support the Stand Up To Cancer charity.
