<\/p>\n
Atrial fibrillation affects more than 58 million people worldwide and significantly increases the risk of stroke. However, current tools for assessing this risk, such as classical scores, remain imprecise and do not account for the complex interactions between various patient health factors.<\/p>\n
A team of researchers has developed new artificial intelligence models capable of predicting the one-year risk of stroke in patients newly diagnosed with atrial fibrillation. These models use only easily accessible information: the patient’s age, medical history, and medications. Unlike traditional methods, these tools analyze the subtle relationships between these elements to provide a more personalized and reliable risk estimate.<\/p>\n
The results are compelling. The two models tested\u2014one based on a classical statistical approach and the other on an advanced machine learning technique\u2014demonstrated a significantly higher ability to identify at-risk patients compared to existing methods. Their performance was validated across diverse patient groups, confirming their reliability in various clinical contexts. Additionally, these models allow for adjusting risk thresholds based on individual needs, which is particularly useful for elderly patients or those with multiple health issues.<\/p>\n
Another major advantage is their ease of use. They do not require blood tests or medical imaging, which are often unavailable at the time of diagnosis. This makes them accessible for everyday hospital practice. The researchers also confirmed that these tools work equally well for both men and women, thus avoiding gender-related biases.<\/p>\n
In the long term, these models also help identify patients who would benefit most from anticoagulant therapy. Data show that patients classified as high-risk by these tools experience a significant reduction in stroke risk when taking these medications, unlike those identified by traditional methods.<\/p>\n
This advancement paves the way for more personalized medicine. Doctors may soon rely on these predictions to tailor treatments based on each patient’s unique profile, thereby reducing the number of preventable strokes and improving the management of atrial fibrillation.<\/p>\n
DOI:<\/strong> https:\/\/doi.org\/10.1038\/s41746-026-02470-3<\/a><\/p>\n Title:<\/strong> Interpretable machine learning models for stroke risk prediction in patients with newly diagnosed atrial fibrillation<\/p>\n Journal:<\/strong> npj Digital Medicine<\/p>\n Publisher:<\/strong> Springer Science and Business Media LLC<\/p>\n Authors:<\/strong> Jesse Chih-Wei Lin; Chen-Min Chang; Heng-Yu Pan; Yi-Lwun Ho; Yu-Kang Tu; Chao-Lun Lai<\/p>\n","protected":false},"excerpt":{"rendered":" Can We Better Predict the Risk of Stroke in Patients with Atrial Fibrillation? Atrial fibrillation affects more than 58 million people worldwide and significantly increases the risk of stroke. However, current tools for assessing this risk, such as classical scores, remain imprecise and do not account for the complex interactions between various patient health factors.… Continue reading Can We Better Predict the Risk of Stroke in Patients with Atrial Fibrillation?<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2,7],"tags":[],"class_list":["post-45","post","type-post","status-publish","format-standard","hentry","category-health","category-international","entry"],"_links":{"self":[{"href":"https:\/\/journalofartificialintelligence.com\/en\/wp-json\/wp\/v2\/posts\/45","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/journalofartificialintelligence.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/journalofartificialintelligence.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/journalofartificialintelligence.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/journalofartificialintelligence.com\/en\/wp-json\/wp\/v2\/comments?post=45"}],"version-history":[{"count":1,"href":"https:\/\/journalofartificialintelligence.com\/en\/wp-json\/wp\/v2\/posts\/45\/revisions"}],"predecessor-version":[{"id":46,"href":"https:\/\/journalofartificialintelligence.com\/en\/wp-json\/wp\/v2\/posts\/45\/revisions\/46"}],"wp:attachment":[{"href":"https:\/\/journalofartificialintelligence.com\/en\/wp-json\/wp\/v2\/media?parent=45"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/journalofartificialintelligence.com\/en\/wp-json\/wp\/v2\/categories?post=45"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/journalofartificialintelligence.com\/en\/wp-json\/wp\/v2\/tags?post=45"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}