Changes in AI mammogram scores may help predict breast cancer years before diagnosis, research involving more than 54,000 women suggests.

Scores rose steadily among women who later developed the disease but remained broadly stable among those who did not.

The increase could be detected up to six years before diagnosis and became much steeper during the final two years.

Researchers led by Professor Constance Lehman, of Harvard Medical School and healthcare technology company Clairity, analysed screening mammograms taken between 2009 and 2019.

They used a validated, open-source deep learning model to calculate five-year breast cancer risk scores from the images alone.

Deep learning is a form of artificial intelligence trained to recognise complex patterns in large amounts of data.

The model examined the whole mammogram rather than relying on a limited, predetermined feature such as breast density.

Models of this kind have performed better than traditional risk models and breast density alone when estimating a woman’s five-year breast cancer risk.

The study initially included 239,703 consecutive two-dimensional screening mammograms from 89,882 patients across six imaging sites spanning urban tertiary, community-based and rural settings.

All were standard bilateral full-field digital mammography examinations, taken with or without digital breast tomosynthesis.

Digital breast tomosynthesis uses multiple low-dose X-ray images to create a three-dimensional view of the breast.

After exclusions, the final analysis involved 54,014 women with a median age of 61 and a total of 158,807 mammograms.

Each woman contributed one index examination and up to six previous annual mammograms. Women had a median of three scans each.

For women who developed cancer, the index examination was their final screening mammogram within the year before diagnosis. For the cancer-free group, it was their final mammogram during the five-year study period.

The model did not use demographic information, clinical records or historical imaging data when calculating each score.

Of the women included, 817, or one per cent, were diagnosed with breast cancer within 365 days of their index examination.

This included 451 women, or 55 per cent, with invasive breast cancer and 118, or 14 per cent, with ductal carcinoma in situ, known as DCIS.

DCIS occurs when abnormal cells are found inside a milk duct but have not spread into the surrounding breast tissue.

The cancer type was unknown for the remaining 248 patients, representing 30 per cent of the cancer group.

A total of 682 cancers, or 83 per cent, were detected through screening, while 135, or 17 per cent, were interval cancers diagnosed between routine mammograms.

The other 53,197 women were not diagnosed with breast cancer during follow-up and formed the cancer-free comparison group.

Professor Lehman said: “We observed clinically relevant differences in risk trajectories between women who did and did not develop cancer. The increase in scores among cancer patients was detectable as early as six years prior to diagnosis and became more pronounced over time.”

Among women later diagnosed with the disease, the median score rose from 2.1 five to six years before diagnosis to 6.6 at the index examination.

Scores among cancer-free women remained stable, with median values ranging from 1.8 to 2.2 throughout the study.

The rise among women who developed cancer was steepest during the two years before their index examination.

Professor Lehman said: “These findings demonstrate signals, invisible to the human eye, in the image alone can predict future risk. This is exciting, because 85 per cent of women diagnosed with breast cancer do not have a significant family history of breast cancer or known genetic mutations.”

Most breast cancers are considered sporadic, meaning they are not driven by inherited genetic changes or a family history of the disease.

Traditional risk models have a limited ability to distinguish between women who will and will not develop breast cancer when used across large screening populations.

Researchers said tracking how scores change over time could provide more information than calculating risk at a single appointment.

Professor Lehman said: “AI-derived risk scores can identify patients who are otherwise predisposed to the disease, and our findings demonstrate that image-based AI risk scores evolve over time and that changes in those scores may provide additional information about future breast cancer risk.”

The patterns remained consistent when women were grouped by age and breast density.

Breast density describes the amount of fibrous and glandular tissue visible on a mammogram. Dense tissue can make cancers harder to detect and is also associated with an increased risk of the disease.

Researchers said image-based scores could support personalised screening and risk-reduction strategies without relying on self-reported or inconsistent clinical information.

Professor Lehman said: “These trends remained robust across subgroups defined by age and breast density, further supporting the generalisability of our findings. This is particularly relevant given persistent disparities in screening performance across patient populations. A dynamic biomarker approach grounded in the imaging data could mitigate some of these disparities by enabling risk-based personalisation that does not rely on self-reported or inconsistent clinical data.”

A biomarker is a measurable sign that can indicate a person’s health, disease risk or response to treatment.

Changing scores could eventually help clinicians identify women who may benefit from additional imaging or measures intended to reduce their risk.

Professor Lehman said: “With the power of AI, computer vision, and the ability to extract predictive data, we are able to apply the power of imaging to risk assessment and preventing disease from developing. Having a dynamic risk score opens up a whole new domain of more effective preventive therapies for breast cancer, similar to how we screen for and treat patients with high cholesterol and hypertension.”

AI image-based risk scores are included in the 2026 National Comprehensive Cancer Network guidelines.

The guidelines recommend that, from the age of 35, women with an elevated five-year risk score of more than 1.7 per cent consider breast MRI alongside annual mammography.

An AI image-based model approved by the US Food and Drug Administration is already being used to calculate five-year breast cancer risk at selected US healthcare institutions.

LGBTQ+ people across the UK face discrimination, funding inequalities and gaps in fertility care, research has found.

Eligibility for NHS-funded treatment varies across the country, while many services are still structured around heterosexual couples.

People with diverse sexual orientations and gender identities can be left navigating complex systems, paying more for treatment and explaining their needs to healthcare professionals.

Co-author Dr Chloe He, of the UCL Institute of Epidemiology and Health Care, said: “Legal access is not the same as equitable access. LGBTQ+ patients are forced to navigate a Kafkaesque fertility care system alone – researching, self-advocating, and often educating the doctors and nurses treating them.

“In our study, we saw clinicians with no formal LGBTQ+ training, gay men pressured into being relentlessly cheerful to prove parent-worthiness to surrogacy services, and patients travelling hundreds of miles for care after experiencing transphobia at local clinics.”

The University of Stirling-led research involved 54 participants and 36 in-depth interviews with people who had used fertility services and professionals working in or alongside fertility care across the UK.

Researchers from Stirling, SKEMA Business School and University College London examined the extra work undertaken by LGBTIQA+ people seeking to have children.

They called this “reproductive labour”, which includes researching treatment, advocating for themselves, covering additional costs and educating clinicians.

The researchers said this work was used to manage “reproductive bioprecarity”, a term describing the uncertainty and vulnerability people can face while seeking reproductive healthcare.

The study, funded by a Santander Universities Research Grant, primarily reflected the experiences of cisgender lesbian participants.

One participant, Amanda, said she and her partner, Amy, spent a long time trying to find a GP willing to discuss fertility with them.

The couple eventually underwent fertility tests through the NHS, but their private clinic rejected the results because they had not been referred by a GP.

They had to repeat the tests and pay for them privately.

The researchers said lesbian couples are often required to self-fund multiple rounds of intrauterine insemination before becoming eligible for NHS support.

Intrauterine insemination, or IUI, involves placing sperm directly into the womb.

Gay men usually have to pursue surrogacy, which is not funded or supported by the NHS, while transgender people can face long waits to save eggs and sperm to allow them to have children.

Lead author Dr Carolyn Wilson-Nash, senior lecturer at the University of Stirling Business School, began investigating the issue after she and her wife made multiple attempts to conceive and faced challenges throughout the process.

The couple funded almost the entire process themselves and consulted a GP who had no experience of supporting same-sex couples seeking fertility care.

The researchers called for clearer treatment pathways, more inclusive services and better training for healthcare staff.

Dr Wilson-Nash, who is now the mother of a three-year-old boy, said: “The way the current system for fertility services is set up in the UK can lead to unequal pathways for the LGBTIQA+ population.

“For example, heterosexual couples can access NHS-funded in vitro fertilisation (IVF), whereas lesbian couples are often required to self-fund multiple rounds of intrauterine insemination (IUI) before becoming eligible for NHS support.

“Gay men usually have to pursue surrogacy, which is not funded by or supported by the NHS.

“And transgender individuals often face long waiting times to save eggs and sperm to allow them to have children. So legal access does not necessarily translate into equitable or inclusive care.

“Building a family should be neither exclusive nor this difficult. Fertility services should be available to all, regardless of their sexual orientation or gender identity.”

Laura-Rose Thorogood, founder of LGBT Mummies and part of the UK’s Fertility Justice Campaign, said: “Right now, intended LGBTQIA+ parents are being discriminated against because of who they are, and who they love.

“This is ultimately forcing them down alternative pathways which in turn put them at long-term risk physically, psychologically and socially.

“By providing access to treatment, our community can thrive and create the families they dream of by their chosen route.”

A sold-out summit at the London Institute for Healthcare Engineering will bring together clinicians, researchers, regulators, investors and founders to discuss how artificial intelligence is being applied to women’s health, and what responsible development of these tools should look like.

AI × Women’s Health: Innovation, Challenges and Opportunities takes place on 25 June.

It’s organised by MEGI Health, a femtech company building a digital cardiovascular platform aimed at supporting women’s heart health through pregnancy, postpartum and beyond.

All 140 places have gone, and there’s now a waiting list.

Nina Sesto, CEO of MEGI Health, said: “We are seeing a wave of innovation in women’s health, and AI has the potential to accelerate it.

“The opportunity is enormous, but it only pays off if these tools are built on representative data and designed around the realities of women’s health.

“That is exactly the conversation we wanted to convene, across clinicians, researchers, industry and regulators.”

AI is increasingly being applied across women’s healthcare, with proponents pointing to earlier diagnosis, better risk prediction and more personalised care.

The summit will look at tools in development across fetal and gynaecological ultrasound, maternal cardiovascular monitoring and clinical decision support, alongside the question of whether women risk being left behind as the technology develops.

The timing reflects wider momentum in the sector: women’s health has been drawing more investment and policy attention, and the World Economic Forum and McKinsey Health Institute have estimated that closing the women’s health gap could add at least $1 trillion a year to the global economy by 2040.

A recurring theme for the day is data.

Women have historically been under-represented in medical research, and organisers argue that AI tools trained on incomplete or unrepresentative datasets risk repeating those gaps rather than closing them.

Sessions on data, ethics, privacy and equity will run alongside the more technical and commercial discussions.

Dr Fran Conti-Ramsden is a clinician at Guy’s and St Thomas’ NHS Foundation Trust, academic at King’s College London and Chief Medical Officer of MEGI Health.

Conti-Ramsden said: “Working at the intersection of clinical practice, academia and industry, I see both the tremendous challenges we face in delivering clinical care for women and the need for innovation, alongside the rapid development of AI and digital health technologies.

“But bringing innovation into clinical practice is fraught with challenges.

“I hope this day brings together people from across the landscape to discuss and define those challenges as well as celebrate progress in the field, sparking dialogue on how we should innovate responsibly, and to make sure women’s health is not left behind.”

The half-day programme is split into four sessions, chaired respectively by Professor Eugene Oteng-Ntim, Professor Richard Dobson, an interactive breakout segment, and Professor Asma Khalil. Other speakers include MiRa Jacobs (MHRA), Professor Jane Hirst (The George Institute), Tulsi Patel (Hertility) and Dr Kimberley Peven (Scarlet), among others.

Organisers say they hope the event will help build a longer-term UK community around clinical AI in women’s health.