Diagnosis
AI-powered mammograms: a new window into heart health
Mammograms used in combination with AI may reveal much more than cancer, but can also be used to assess the amount of calcium build up in the arteries within breast tissue – an indicator of cardiovascular health, a new study shows.
While breast artery calcifications can be seen on mammogram images, radiologists do not typically quantify or report this information to women or their clinicians.
This new study, which used an AI image analysis technique not previously used on mammograms, demonstrates how AI can help fill this gap by automatically analysing breast arterial calcification and translating the results into a cardiovascular risk score.
“We see an opportunity for women to get screened for cancer and also additionally get a cardiovascular screen from their mammograms,” said Theo Dapamede, postdoctoral fellow at Emory University in Atlanta and the study’s lead author.
“Our study showed that breast arterial calcification is a good predictor for cardiovascular disease, especially in patients younger than age 60. If we are able to screen and identify these patients early, we can refer them to a cardiologist for further risk assessment.”
Heart disease is the leading cause of death in the United States but remains underdiagnosed in women and there is also lagging awareness. Researchers said the use of AI-enabled mammogram screening tools could help identify more women with early signs of cardiovascular disease by taking better advantage of screening tests that many women routinely receive.
A build up of calcium in blood vessels is a sign of cardiovascular damage associated with early-stage heart disease or aging. Previous studies have shown that women with calcium build up in the arteries face a 51 per cent higher risk of heart disease and stroke.
To develop the screening tool used for this study, researchers trained a deep-learning AI model to segment calcified vessels in mammogram images, which appear as bright pixels on X-rays, and calculate the future risk of cardiovascular events based on data obtained from the electronic health record data.
The segmentation approach is what separates this model from previous AI models developed for analysing breast artery calcifications. Researchers said the model is also strengthened by its use of a large dataset for training and testing, which included images and health records from over 56,000 patients who had a mammogram at Emory Healthcare between 2013 and 2020 and had at least five years of follow-up electronic health records data.
“Advances in deep learning and AI have made it much more feasible to extract and use more information from images to inform opportunistic screening,” Dapamede said.
Overall findings showed the new model performed well at characterizing patients’ cardiovascular risk as low, moderate or severe based on mammogram images.
After calculating the risk of dying from any cause or suffering an acute heart attack, stroke or heart failure at two years and five years, the model showed that the rate of these serious cardiovascular events increased with breast arterial calcification level in two of the three age categories assessed – women younger than age 60 and age 60 to 80, but not in those over age 80.
This makes the tool particularly well suited for providing early warning of heart disease risk in younger women, who can benefit more from early interventions, researchers said.
The results also showed that women with the highest level of breast arterial calcification (above 40 mm2) had a significantly lower five-year rate of event-free survival than those with the lowest level (below 10 mm2).
For example, 86.4 per cent of those with the highest breast arterial calcification survived for five years compared with 95.3 per cent of those with the lowest level of calcification. This translates to approximately 2.8 times the risk of death within five years in patients with severe breast arterial calcification compared to those with little to no breast arterial calcification.
The AI model was developed as a collaboration between Emory Healthcare and Mayo Clinic and is not currently available for use.
If it passes external validation and gains approval from the U.S. Food and Drug Administration, researchers said the tool could be made commercially available for other health care systems to incorporate into routine mammogram processing and follow-up care.
The researchers also plan to explore how similar AI models could be used for assessing biomarkers for other conditions, such as peripheral artery disease and kidney disease, that might be extracted from mammograms.
AI could help routine mammograms spot heart disease risk in women, as well as breast cancer, researchers have found.
A team from Emory University analysed regular screening scans from more than 123,000 women who had no prior history of cardiovascular disease.
Using an AI programme to quantify calcification and hardening in the arteries, they found women with severe cases had two to three times the risk of developing potentially fatal heart attacks, stroke and heart failure.
“This was true even in younger women under 50, a group often considered low-risk, and held up after accounting for other risk factors like diabetes and smoking,” said Hari Trivedi, Emory associate professor and co-director of the university’s Health Innovation and Translational Informatics lab, who described the research as the largest study of its kind.
“We wanted to test whether AI could use this to identify women at risk of cardiovascular disease at no extra cost or inconvenience.
“”For women, this means a mammogram you’re already having could also provide important information about your heart health, prompting a conversation with your doctor about preventive steps such as cholesterol testing or medication.”
The researchers’ work was published today in the European Heart Journal, from the European Society of Cardiology, which noted that women are largely underdiagnosed and undertreated for cardiovascular disease.
They wrote that the high resolution of routine mammograms could allow for essentially automatic, direct visualisation of the arterial beds within the breasts of nearly all adult women, with calcifications easy to detect and correlated with deposits found in other parts of the body.
The researchers said that, compared with imaging of the heart’s coronary arteries, where blockages narrow and impede the flow of oxygenated blood to the heart muscle, calcifications in breast tissue affect a separate layer of the vessel, resulting in increased artery stiffness, a measure typically linked to long-term hypertension.
They described the finding as an independent predictor of cardiovascular disease and a potentially useful addition to traditional cardiovascular risk factors.
In an accompanying editorial, Lori Daniels, a cardiologist and professor of medicine at the UC San Diego School of Medicine, said that while fewer than 40 per cent of women may know their cholesterol levels, many more are up to date with their breast cancer screening.
“Two-thirds of women aged 50-69 in the European Union reported a mammogram within the prior 2 years, and in the USA, nearly 70 per cent of women aged 45 years and older were up to date with mammography according to American Cancer Society screening guidelines,” Daniels wrote.
“Breast arterial calcification has the potential to reframe this mismatch, leveraging a widely adopted cancer-screening platform to identify cardiovascular risk in women who may not otherwise engage with prevention.”
A blood test for endometriosis showed clinical promise after detecting cases missed by standard imaging, according to a clinical validation study.
HerAnova Lifesciences has published a peer-reviewed clinical validation study of its HerResolve blood test for endometriosis in the Journal of Minimally Invasive Gynecology, the official journal of the AAGL.
The multi-centre study enrolled 298 women of reproductive age across 11 clinical sites in the US, Europe and Hong Kong.
The study population was 75.8 per cent white, 9.7 per cent Black, 9.1 per cent Asian and 5 per cent non-white Hispanic participants.
It found the test identified 61.5 per cent of histologically confirmed endometriosis cases that were missed by transvaginal ultrasound and or MRI scans.
All results were validated against the gold standard of laparoscopic findings with histopathological tissue confirmation.
The headline numbers were an AUC of 0.944, specificity of 97.5 per cent and sensitivity of 80 per cent. The high specificity was a deliberate design choice, with the model optimised to minimise false positives and reduce unnecessary invasive procedures. Performance was also consistent across menstrual phases.
The blood test, called HerResolve, is a multi-omic blood-based assay that combines three serum microRNA biomarkers, three protein biomarkers, one steroid hormone, patient age and BMI into a machine learning algorithm to detect endometriosis.
Farideh Bischoff, chief medical officer at HerAnova and corresponding author of the study, said: “Endometriosis has long been one of the most underdiagnosed and undertreated conditions in women’s health.
“HerResolve was designed to work alongside existing imaging and clinical evaluation, filling a critical gap in non-invasive disease detection.”
The test is currently available at select IVF and reproductive medicine centres across the US and is positioned as a triage tool, helping identify patients who may benefit from further evaluation or empirical treatment rather than replacing surgery entirely, but potentially reserving it for treatment rather than diagnosis.
A prospective validation study is underway in geographically and ethnically diverse populations, and HerAnova is also pursuing longitudinal analyses to evaluate whether the assay can monitor treatment response over time.
Endometriosis affects approximately one in 10 women of reproductive age, yet the average diagnostic delay remains six to 11 years.
The current gold standard, laparoscopic surgery, is invasive, dependent on surgeon skill and not without risk, making a reliable non-invasive alternative one of the most sought-after tools in women’s health diagnostics.
New research has revealed privacy and advice concerns surrounding tech aimed at helping women navigate the menopause.
The study, which surveyed 310 UK participants, warns that sensitive data collected from women about their health is often vulnerable to exploitation, ranging from targeted financial scams to workplace discrimination and misinformation.
With nearly one million women in the UK leaving their jobs due to menopausal symptoms, and an estimated 13m currently perimenopausal or menopausal, new digital technology has boomed, promising to provide medical advice and solutions.
However, the research identified significant security gaps in these technologies, including the lack of medical professionals on community forums, leading to the spread of unverified medical advice and misinformation.
The large-scale study, which is the first of its kind, also highlighted a growing tension between the benefits of menopause tech and the privacy risks associated with the intimate data these services collect.
Unlike fertility trackers, which have faced intense scrutiny following legal changes such as the overturning of Roe v. Wade in the US, menopause tech remains under-researched and under-regulated.
Dr Maryam Mehrnezhad, from the information security department at Royal Holloway and co-author, said: “Many women are turning to technology to navigate the diagnostic ambiguity of going through the menopause and to combat the long waiting times often associated with traditional healthcare.”
This makes the absence of trained medical professionals on these tech community forums, who can create misinformation, a real serious threat to participants.
Users we surveyed also expressed deep fears regarding data use, specifically that intimate health data, including emotional symptoms and sexual activity history, could be accessed by insurance companies or employers.
Such data types can be used to discriminate users e.g., in regarding their health insurances and in workplaces.
Dr Taylor Robinson, co-author and post-doctoral researcher at Royal Holloway, added: “Self-tracking apps are becoming essential tools for personal advocacy, allowing users to document their journeys and foster deeper self-reflection, which ultimately improves their interactions with doctors, but more needs to be done to protect those using the apps.”
As the femtech industry is projected to reach nearly US$30bn by 2032, the authors argue that developers and policymakers must prioritise privacy to ensure digital tools remain a safe resource for those experiencing the complexities of menopause.
Rebecca Jones and Sophie Hawkes, PhD students and co-authors, added that digital platforms and social media groups provide a vital lifeline by alleviating the isolation often tied to menopause, offering a dedicated space for users to share advice and validate their experiences and, with much more rigorous scrutiny on real healthcare advice, they can be a great asset for many.
This research provides crucial practical data to inform secure, user-centric design for policymakers and menopause tech developers, noting that many current apps and devices fall short of GDPR standards by making privacy notices difficult to access or understand.
Entrepreneur3 weeks agoOura launches women’s health AI model
Insight2 weeks agoThe missing layer in the women’s health conversation
Insight3 weeks agoWomen’s HealthX launches in Boston this December to transform women’s health through data and science
International Women's Day 20262 weeks agoHow AI is shaping the future of women’s health
Entrepreneur2 weeks agoMenstrual health apps market tipped to reach US$13bn
Entrepreneur4 weeks agoMatresa raises £315k for maternal health platform
pain conditions4 weeks agoEndometriosis-sufferer wins landmark tribunal case on workplace discrimination
Motherhood4 weeks agoState abortion bans linked to spike in maternal deaths, study finds