AI spots deadly heart risk most doctors can't see
A new AI model is much better than doctors at identifying patients likely to experience cardiac arrest.
The linchpin is the system's ability to analyze long-underused heart imaging, alongside a full spectrum of medical records, to reveal previously hidden information about a patient's heart health.
The federally-funded work, led by Johns Hopkins University researchers, could save many lives and also spare many people unnecessary medical interventions, including the implantation of unneeded defibrillators.
"Currently we have patients dying in the prime of their life because they aren't protected and others who are putting up with defibrillators for the rest of their lives with no benefit," said senior author Natalia Trayanova, a researcher focused on using artificial intelligence in cardiology. "We have the ability to predict with very high accuracy whether a patient is at very high risk for sudden cardiac death or not."
The findings are published today in Nature Cardiovascular Research.
Hypertrophic cardiomyopathy is one of the most common inherited heart diseases, affecting one in every 200 to 500 individuals worldwide, and is a leading cause of sudden cardiac death in young people and athletes.
Many patients with hypertrophic cardiomyopathy will live normal lives, but a percentage are at significant increased risk for sudden cardiac death. It's been nearly impossible for doctors to determine who those patients are.
Current clinical guidelines used by doctors across the United States and Europe to identify the patients most at risk for fatal heart attacks have about a 50% chance of identifying the right patients, "not much better than throwing dice," Trayanova says.
The team's model significantly outperformed clinical guidelines across all demographics.
Multimodal AI for ventricular Arrhythmia Risk Stratification (MAARS), predicts individual patients' risk for sudden cardiac death by analyzing a variety of medical data and records, and, for the first time, exploring all the information contained in the contrast-enhanced MRI images of the patient's heart.
People with hypertrophic cardiomyopathy develop fibrosis, or scarring, across their heart and it's the scarring that elevates their risk of sudden cardiac death. While doctors haven't been able to make sense of the raw MRI images, the AI model zeroed right in on the critical scarring patterns.
"People have not used deep learning on those images," Trayanova said. "We are able to extract this hidden information in the images that is not usually accounted for."
The team tested the model against real patients treated with the traditional clinical guidelines at Johns Hopkins Hospital and Sanger Heart & Vascular Institute in North Carolina.
Compared to the clinical guidelines that were accurate about half the time, the AI model was 89% accurate across all patients and, critically, 93% accurate for people 40 to 60 years old, the population among hypertrophic cardiomyopathy patients most at-risk for sudden cardiac death.
The AI model also can describe why patients are high risk so that doctors can tailor a medical plan to fit their specific needs.
"Our study demonstrates that the AI model significantly enhances our ability to predict those at highest risk compared to our current algorithms and thus has the power to transform clinical care," says co-author Jonathan Crispin, a Johns Hopkins cardiologist.
In 2022, Trayanova's team created a different multi-modal AI model that offered personalized survival assessment for patients with infarcts, predicting if and when someone would die of cardiac arrest.
The team plans to further test the new model on more patients and expand the new algorithm to use with other types of heart diseases, including cardiac sarcoidosis and arrhythmogenic right ventricular cardiomyopathy.
Authors include Changxin Lai, Minglang Yin, Eugene G. Kholmovski, Dan M. Popescu, Edem Binka, Stefan L. Zimmerman, Allison G. Hays, all of Johns Hopkins; Dai-Yin Luand M. Roselle Abrahamof the Hypertrophic Cardiomyopathy Center of Excellence at University of California San Francisco; and Erica Schererand Dermot M. Phelanof Atrium Health.
Even low levels of air pollution may quietly scar your heart, MRI study finds
Researchers using cardiac MRI have found that long-term exposure to air pollution is associated with early signs of heart damage, according to a study that was published today in Radiology, a journal of the Radiological Society of North America (RSNA). The research indicates that fine particulate matter in the air may contribute to diffuse myocardial fibrosis, a form of scarring in the heart muscle that can precede heart failure.
Cardiovascular disease is the leading cause of death worldwide. There is a large body of evidence linking poor air quality with cardiovascular disease. However, the underlying changes in the heart resulting from air pollution exposure are unclear.
"We know that if you're exposed to air pollution, you're at higher risk of cardiac disease, including higher risk of having a heart attack," said the study's senior author Kate Hanneman, M.D., M.P.H., from the Department of Medical Imaging at the Temerty Faculty of Medicine, University of Toronto and University Health Network in Toronto. "We wanted to understand what drives this increased risk at the tissue level."
Dr. Hanneman and colleagues used cardiac MRI, a noninvasive imaging technique, to quantify myocardial fibrosis and assess its association with long-term exposure to particles known as PM2.5. At 2.5 micrometers in diameter or less, PM2.5 particles are small enough to enter the bloodstream through the lungs. Common sources include vehicle exhaust, industrial emissions and wildfire smoke.
The researchers wanted to evaluate the effects of air pollution on both healthy people and those with heart disease, so the study group included 201 healthy controls and 493 patients with dilated cardiomyopathy, a disease that makes it more difficult for the heart to pump blood.
Higher long-term exposure to fine particulate air pollution was linked with higher levels of myocardial fibrosis in both the patients with cardiomyopathy and the controls, suggesting that myocardial fibrosis may be an underlying mechanism by which air pollution leads to cardiovascular complications. The largest effects were seen in women, smokers and patients with hypertension.
The study adds to growing evidence that air pollution is a cardiovascular risk factor, contributing to residual risk not accounted for by conventional clinical predictors such as smoking or hypertension.
"Even modest increases in air pollution levels appear to have measurable effects on the heart," Dr. Hanneman said. "Our study suggests that air quality may play a significant role in changes to heart structure, potentially setting the stage for future cardiovascular disease."
Knowing a patient's long-term air pollution exposure history could help refine heart disease risk assessment and address the health inequities that air pollution contributes to both in level of exposure and effect. For instance, Dr. Hanneman said, if an individual works outside in an area with poor air quality, healthcare providers could incorporate that exposure history into heart disease risk assessment.
The air pollution exposure levels of the patients in the study were below many of the global air quality guidelines, reinforcing that there are no safe exposure limits.
"Public health measures are needed to further reduce long-term air pollution exposure," Dr. Hanneman said. "There have been improvements in air quality over the past decade, both in Canada and the United States, but we still have a long way to go."
In addition to illuminating the links between air pollution and myocardial fibrosis, the study highlights the important role that radiologists will play in research and clinical developments going forward.
"Medical imaging can be used as a tool to understand environmental effects on a patient's health," Dr. Hanneman said. "As radiologists, we have a tremendous opportunity to use imaging to identify and quantify some of the health effects of environmental exposures in various organ systems."
Sweet-smelling molecule halts therapy-resistant pancreatic cancer
Cancer cells have the capacity to multiply rapidly. The aggressive cancer cells undergo conversion from their tightly connected epithelial state into a mesenchymal state, which lacks contact restrictions and spreads easily to other parts of the body. Such epithelial-to-mesenchymal plasticity also makes the cancer cells resistant to elimination by anticancer therapies.
The search is ongoing for newer anticancer agents that can overcome this acquired resistance to therapy and destroy the 'rogue' cancer cells. A group of researchers led by Dr. Hideyuki Saya, Director of the Oncology Innovation Center, Fujita Health University, Japan, has uncovered the mechanism of the anticancer activity of benzaldehyde, a compound responsible for the aroma of almonds, apricots, and figs.
Giving insights into their motivation for this study, Dr. Saya explains, "In the 1980s, researchers demonstrated the anticancer activity of benzaldehyde and its derivatives. The first author of our study, Dr. Jun Saito, is the daughter of one of the researchers involved in those early studies, and she was driven by a strong desire to uncover the mechanism behind benzaldehyde's anticancer effects." This study, published online in the British Journal of Cancer on May 02, 2025, shows the impact of benzaldehyde on key signaling protein interactions within the cancer cells and the resulting cytotoxicity.
Early studies reported the ability of benzaldehyde to inhibit the progressive development of mouse embryonic cells, indicating its potential in preventing rapid cell proliferation. Here, the anticancer effects of benzaldehyde were studied by using a mouse model grafted to have a growing pancreatic cancer.
In cell culture studies, benzaldehyde inhibited the growth of cancer cells resistant to radiation therapy and also those resistant to treatment with osimertinib, an agent blocking tyrosine kinases in growth factor signaling. Benzaldehyde synergized with radiation to eliminate previously radiation-resistant cancer cells.
The study findings revealed that benzaldehyde exerted its anticancer effects by preventing interactions of the signaling protein 14-3-3ζ with the Ser28-phosphorylated form of histone H3 (H3S28ph). This interaction, key to cancer cell survival, was also responsible for treatment resistance and the expression of genes related to epithelial-mesenchymal plasticity.
Here, benzaldehyde prevented 14-3-3ζ-dependent phosphorylation of the serine28 amino acid of histone H3. Consequently, benzaldehyde treatment reduced the expression of genes responsible for treatment resistance. Treatment of mice with a benzaldehyde derivative inhibited the growth of pancreatic tumors and suppressed the epithelial-to-mesenchymal plasticity, thus preventing the spread of cancer to distant organs like the lungs.
By blocking an interaction key to cancer cell survival, benzaldehyde overcomes therapy resistance and prevents metastasis. Sharing the implications of their findings, Dr. Saya concludes, "The 14-3-3ζ protein has long been considered a target for cancer therapy, but its direct inhibition is not feasible due to its important functions in normal cells. Our results suggest that inhibition of the interaction between 14-3-3ζ and its client proteins by benzaldehyde has the potential to overcome the problem."
The present study shows benzaldehyde is effective against cancer cells that have acquired resistance to radiation and tyrosine kinase inhibitors commonly used in cancer treatment. In the long term, this study suggests its potential as a combinatorial anticancer agent, alongside molecular-targeted therapies.
Dr. Jun Saito is a researcher in the laboratory of Dr. Hideyuki Saya, Oncology Innovation Center, Fujita Health University. She obtained her Ph.D. from the Nihon University Graduate School of Medicine. Continuing the legacy of one of her parents, who pioneered the breakthrough research on the anticancer activity of benzaldehyde in the 1980s, Dr. Saito has uncovered the underlying mechanism of benzaldehyde's anticancer effects. Her expertise includes oncology, pathophysiology, immunology, applied physics, and chemistry.
This work was supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (KAKENHI 19K22568).