Here are the top medical news for the day:
Study Reveals How Green Spaces May Protect Children Against ADHD and Autism
A new study published in Environment International by Rutgers Health researchers has found that living near green spaces before and during pregnancy, as well as in early childhood, is associated with a reduced risk of neurodevelopmental disorders. These include attention-deficit/hyperactivity disorder (ADHD), autism spectrum disorder (ASD), and other developmental delays.
Researchers sought to address an important knowledge gap regarding the impact of green space, especially among socioeconomically disadvantaged groups. To investigate, researchers analyzed data from the Medicaid Analytic Extract spanning 2001 to 2014. The dataset included over 1.8 million racially and socioeconomically diverse mother–child pairs across multiple U.S. states. Green space exposure was quantified using satellite imaging to assess vegetation density near mothers’ residential ZIP codes during preconception, pregnancy, and early childhood.
A new study published in HemaSphere by researchers at Queen Mary University of London has shown that molecules exhaled in the breath may help detect blood cancer. The findings could pave the way for the development of a breathalyser-style diagnostic tool that offers a rapid, low-cost, and non-invasive way to identify blood cancers.
Conventional diagnostic methods for cancer diagnosis such as imaging scans or biopsies are expensive and not easily accessible in many parts of the world. A portable breath test could help overcome these limitations and make early detection more widely available.
To explore this possibility, Dr. John Riches and his team at the Barts Cancer Institute used Breath Biopsy®, a breathalyser technology developed by Owlstone Medical. They collected and analyzed exhaled breath from 46 patients with blood cancer and 28 healthy individuals. By using mass spectrometry, the researchers were able to detect tens of thousands of molecular fragments and compare their chemical profiles.
The team discovered that individuals with high-grade lymphoma, an aggressive type of blood cancer, exhaled elevated levels of specific molecules linked to oxidative stress a process involved in cancer development. The researchers suggest this breath-based method could be used to diagnose and monitor blood cancer, particularly in underserved or rural settings.
The team plans to further investigate which lymphoma types are most detectable via breath and refine the method to reduce breath collection time from 10 minutes to just a few seconds.
“Previous studies have shown the value of using breath tests to detect lung cancer. But no one had ever investigated whether blood cancer cells release molecules that pass into the breath, despite the purpose of breathing being to exchange substances between the blood and the breath,” commented Dr. John Riches.
“In future, rather than sending patients away for costly scans and waiting for test results, doctors may be able to conduct a quick breath test in their clinic room and potentially have the results within a few seconds,” said Dr. Riches.
Reference: Stiekema, L.C.A., Chou, H., Craster, A., Wrench, B., Bianchi, K., Gallipoli, P., Davies, J.K., Gribben, J.G. and Riches, J.C. (2025), Analysis of volatile organic compounds in exhaled breath of blood cancer patients identifies products of lipid peroxidation as biomarkers for lymphoma detection. HemaSphere, 9: e70168. https://doi.org/10.1002/hem3.70168
AI ECG Tool Detects Hidden Heart Disease More Accurately Than Doctors: Study Finds
A recent study published in Nature has demonstrated that an artificial intelligence (AI) model can reliably detect diverse structural heart diseases from electrocardiograms (ECGs), outperforming standard physician review. Developed by researchers across eight NewYork-Presbyterian hospitals, the AI model named EchoNext was designed to act as a multitask classifier capable of identifying subtle signs of structural heart diseases that may otherwise go undiagnosed.
Diagnosis of structural heart diseases often relies on expensive and inaccessible tools like echocardiography. Early detection is critical, but over 6% of older adults with significant valvular heart disease remain undiagnosed. EchoNext aims to fill this gap by providing a low-cost, scalable solution through AI interpretation of routine 10-second ECGs.
To develop and validate the model, researchers compiled over 1.2 million ECG-echocardiogram pairs from 230,318 adults treated between 2008 and 2022. EchoNext processed raw 12-lead ECG waveforms, seven standard ECG parameters, and demographic data. It was evaluated using a held-out internal test set as well as external datasets from Cedars-Sinai, the Montreal Heart Institute, and the University of California, San Francisco.
The model performed strongly, detecting composite structural heart diseases with an AUROC of 85.2% internally and 78–80% externally. In a reader study, EchoNext outperformed 13 cardiologists who reviewed de-identified ECGs, achieving 77% accuracy versus 64% for clinicians. When doctors were shown EchoNext's risk score, their accuracy rose to 69%.
The researchers also conducted a silent deployment on over 84,000 imaging-naive patients in 2023, finding that EchoNext could have flagged nearly 2,000 hidden structural heart diseases cases that were otherwise missed. In a small prospective pilot, 73% of high-risk participants had previously unrecognized structural heart diseases.
These findings suggest that AI-based ECG analysis like EchoNext could serve as a valuable screening tool to triage patients, optimize echocardiography use, and improve early structural heart diseases detection, especially in resource-limited settings.
Reference: Poterucha, T.J., Jing, L., Ricart, R.P. et al. Detecting structural heart disease from electrocardiograms using AI. Nature (2025). https://doi.org/10.1038/s41586-025-09227-0
Disclaimer: This website is primarily for healthcare professionals. The content here does not replace medical advice and should not be used as medical, diagnostic, endorsement, treatment, or prescription advice. Medical science evolves rapidly, and we strive to keep our information current. If you find any discrepancies, please contact us at corrections@medicaldialogues.in. Read our Correction Policy here. Nothing here should be used as a substitute for medical advice, diagnosis, or treatment. We do not endorse any healthcare advice that contradicts a physician's guidance. Use of this site is subject to our Terms of Use, Privacy Policy, and Advertisement Policy. For more details, read our Full Disclaimer here.
NOTE: Join us in combating medical misinformation. If you encounter a questionable health, medical, or medical education claim, email us at factcheck@medicaldialogues.in for evaluation.