As the UK is in the midst of its “worst cardiac care crisis in living memory”, new research suggests artificial intelligence could help with earlier diagnosis of people at risk of heart failure.
Thursday 30 May 2024 01:57, UK
Artificial intelligence can identify abnormalities that were previously hard to detect and could play a crucial role in early diagnosis of people at risk of heart failure, according to a new study.
Heart and circulatory disease is the world's leading cause of death. One in three people lose their lives every year. Researchers in Scotland set out to test whether AI could deliver “real-world benefits” to people at risk.
Thanks to patients who voluntarily provided their data to the Scottish Health Research Registry and Biobank (SHARE), researchers from the University of Dundee's School of Medicine studied a final cohort of 578 people to explore how AI might help.
The research team, who published their findings in the journal ESC Heart Failure, used AI to examine population-based electronic health records and cardiac ultrasound scans to identify patients with heart failure.
read more:
Can Ozempic save you from a heart attack?
Intermittent fasting may significantly increase risk of death from cardiovascular disease
It then used AI deep learning to examine the images and identify abnormalities that may pose increased risk to patients.
Professor Chim Lan said: “Our work represents an advancement in the use of deep learning to automatically interpret echocardiographic images.”
“This will enable us to streamline the identification of heart failure patients at scale within electronic health record datasets.
“Echocardiogram heart scans enhanced by AI software have helped provide further measurements, or parameters, of cardiac structure and function that can aid in the diagnosis of heart failure.”
“These measurements were not routinely reported on regular cardiac scans from electronic health records.
“Compared to reports generated from regular cardiac scans, AI-enhanced reports are more detailed and can also be processed at a larger scale than traditional images.
“This could potentially impact clinical practice and research as it could increase the efficiency and speed of patient selection for pragmatic clinical trials and improve surveillance and early diagnosis of heart failure across hospital systems.”
Experts say heart failure, which refers to the inability of the heart to effectively pump blood around the body, is a very common but underdiagnosed condition.
Symptoms can be controlled to some extent with lifestyle changes, surgery and medication, but in most cases it is a serious, long-term condition that gradually worsens over time.
“Real World Benefits”
According to a January report from the British Heart Foundation (BHF), an estimated 20.5 million people will die from heart and circulatory disease worldwide in 2021, which equates to one death every 1.5 seconds.
At the time, Dr Sonia Babu Narayan, deputy medical director of the BHF, said: NHS in the midst of 'worst heart care crisis in living memory' Premature deaths from heart and circulatory disease have risen to their highest level in more than a decade.
Prof Lang said the latest research was “an example of how AI has the potential to deliver real benefits to patients”.
“By evaluating a large number of patient records, we were able to detect structural and functional abnormalities that could not be detected by analysis of traditional echocardiographic images,” he said.
“Although this is a test case, we are excited to be able to apply deep learning on a large scale to Biobank resources.
“We hope this finding will pave the way for other researchers to harness this technology to benefit patients around the world.”
