A pioneering study led by the University of Aberdeen and NHS Grampian has shown that AI-assisted breast cancer screening can detect 10.4% more cancers, reduce radiologist workload by more than 30%, and shorten the time between scan and follow-up notification from 14 days to three — findings now informing the UK-wide EDITH trial and built around real-world screening of 10,889 women in the north-east of Scotland.
The GEMINI study — Grampian’s Evaluation of Mia in an Innovative National breast screening Initiative — was published in Nature Cancer on 10 March 2026 and is the UK’s first comprehensive evaluation of AI in breast cancer screening. Carried out by a team from the University of Aberdeen, NHS Grampian and Kheiron Medical Technologies (now part of DeepHealth Inc.), and facilitated by the North of Scotland NHS Innovation Hub, the study evaluated Kheiron’s “Mia” AI software across 17 different deployment scenarios involving 10,889 women routinely screened in NHS Grampian.
The trial was funded through the NHS AI in Health and Care Award in partnership with the National Institute for Health and Care Research (NIHR) — part of the wider £150 million AI framework now powering NHS innovation across the UK — with project lead Professor Gerald Lip receiving additional support through the Scottish Government’s Chief Scientist Office Innovation Fellowship Programme. The lead author on the published paper is Dr Clarisse de Vries, now a Lecturer in Data Science at the University of Glasgow and a former Research Fellow at Aberdeen.
The headline findings are striking. The optimal Mia configuration — AI as a second reader substituting for one of the two human radiologists, plus an extra AI safeguard pass — increased early cancer detection by 10.4% while reducing human reader workload by more than 30%, without recalling any additional women for follow-up. For women whose scans were flagged for further investigation, time-to-notification fell from an average of 14 days to three.
The case for change set out in the paper is uncomfortable. Dr de Vries said: “As part of the UK breast screening programme all women aged between 50 and 70 years old in the UK are invited for mammograms every three years. This results in over 2 million mammogram examinations being performed annually.”
She continued: “Currently, in the UK, to reduce the number of cancers missed, two radiologists read every mammogram. However, some breast cancers are extremely hard to detect, and it is not always clear from mammograms whether breast cancer is present… Despite this, approximately 20% of cancers are missed using this process. Furthermore, many more women are recalled for further assessments than are diagnosed with cancer. For each five women recalled, approximately one will be diagnosed with breast cancer. So, they have had unnecessary, often invasive tests — not to mention the additional worry for the patient.”
That dual problem — cancers missed and women recalled unnecessarily — is what GEMINI was designed to address. Dr de Vries said: “This is why our findings are so important — not only did we find optimal ways to detect breast cancer, quicker and more accurately, we also found ways to reduce the number of women having to return for unnecessary tests.”
Professor Gerald Lip, Clinical Director for breast screening in the North East of Scotland at NHS Grampian and Lead for Artificial Intelligence in Clinical Practice at the University of Aberdeen, framed the operational implication for the NHS workforce. He said: “Our results show that AI could effectively support breast screening services by increasing cancer detection and reducing doctors’ workload. Ultimately, for radiologists, AI augments practice. Along with picking up more cancers, in UK and European screening programs where mammograms are read by two humans, partial substitution of one of the human readers for normal examinations can deliver real workload savings and reduce burnout. The bottom line here is — without AI, doctors would not have caught these cancers as early.”
Professor Lesley Anderson, Interdisciplinary Institute Director for Health, Nutrition and Wellbeing and Chair in Health Data Science at the University of Aberdeen, set out the methodological significance — work that builds on a track record of Scottish companies advancing medical imaging. She said: “Our unique trial design lets us simulate real-world use of AI in multiple ways, something never done before in this field. This pioneering approach allows healthcare service providers and policymakers to understand better how AI could be operationally integrated into clinical workflows to support breast screening and provide services with different options depending on their needs.”
The findings directly address evidence gaps previously flagged by the UK National Screening Committee, which has not until now recommended AI use in the NHS breast screening programme on the grounds that the evidence base was insufficient. Dr de Vries said: “Our work adds high-quality evidence to the scientific literature in support of AI. It also demonstrates that AI use can be tailored to local healthcare needs to enhance service delivery.”
That tailored, governance-led framing matters. Concerns raised by MPs about poor data quality and legacy systems undermining public-sector AI adoption have been a recurring brake on NHS AI deployment, and Scottish ministers have called for mandatory registration of AI usage across the public sector precisely to ensure such deployment is visible and accountable. GEMINI is, in effect, a worked example of how to do prospective NHS AI evaluation properly.
GEMINI’s findings now feed directly into the UK-wide EDITH trial (Evaluating the use of AI in Detecting cancers wITH mammography), with the Scottish element led jointly by the University of Aberdeen, NHS Grampian and the University of Glasgow. EDITH represents one of the most significant Scotland-led contributions to UK medical AI evaluation in recent years and positions Grampian as an international reference site for AI screening governance — and is part of a broader push that has already seen 10,000 NHS Scotland patients benefit from robotic-assisted surgery and a growing pipeline of clinical innovation supported by the NHS Clinical Entrepreneur Programme and the country’s wider life sciences base, which now includes cancer-focused work from companies such as Cumulus Oncology.
For Scotland’s wider AI-in-government and digital-health agenda, GEMINI is a concrete demonstration of what high-quality, prospectively-evaluated, clinically-governed AI deployment in the NHS actually looks like. The methodology — large-scale prospective NHS evaluation, transparent governance, force-multiplier rather than replacement framing, real workload measurement — provides a template that other Scottish AI-in-public-services programmes are likely to draw on directly.
Source(s): University of Aberdeen — Pioneering study finds AI increases cancer detection by more than 10 percent · Nature Cancer — Prospective evaluation of artificial intelligence integration in breast screening · BBC News — Breast cancer detection ‘up by 10% with use of AI’
