AI Tools Can Support Early Cardiovascular Disease Diagnosis

Recent research from Rutgers, the State University of New Jersey, described how artificial intelligence (AI) and machine learning (ML) could be used to examine genes within DNA, information that can help clinicians predict cardiovascular diseases such as atrial fibrillation and heart failure.

According to the Centers for Disease Control and Prevention (CDC), heart disease is the leading cause of death for people of most ethnic groups in the US, claiming the life of one person every 34 seconds.

In addition, the press release noted that around 50 percent of patients with cardiovascular disease die within five years following a diagnosis, often due to genetics or environmental factors.

However, the World Health Organization also estimated that over 75 percent of premature cardiovascular disease is preventable, according to the Rutgers study published in Genomics.

“Timely understanding and precise treatment of cardiovascular disease will ultimately benefit millions of individuals by reducing the high risk for mortality and improving the quality of life,” said Zeeshan Ahmed, PhD, a core faculty member at the Rutgers Institute for Health, Health Care Policy and Aging Research (IFH) and lead author of the study, in a press release.

To advance early diagnosis and treatment of cardiovascular disease, Rutgers researchers used AI and ML models to identify genes that may have a relationship with cardiovascular disease.

The researchers analyzed healthy patients and patients with cardiovascular disease. They used AI and ML models to determine the association between genes and various manifestations of the disease, including atrial fibrillation and heart failure.

They identified a group of genes associated with cardiovascular disease. They also found that age, gender, and race factors contributed to cardiovascular disease. Specifically, age and gender correlated to heart failure, and age and race were more prominent factors in atrial fibrillation.

“With the successful execution of our model, we predicted the association of highly significant cardiovascular disease genes tied to demographic variables like race, gender and age,” said Ahmed in a press release.  

Researchers noted that further research should review complete sets of genes in patients with cardiovascular disease, as this information can provide valuable information on biomarkers and risk factors.

This is the latest example of AI being used to enhance clinical care.

A study from September 2022 found that an AI-based screening strategy could use electrocardiogram data to successfully determine the risk of a stroke.

After analyzing 1,003 patients who participated in continuous monitoring and 1,003 patients who received standard care, researchers found that an AI algorithm detected atrial fibrillation in six of the 370 low-risk patients and 48 of 633 high-risk patients.

From these findings, researchers concluded that the AI-guided screening tool effectively increased atrial fibrillation detection.

Other research from April 2022 found that an AI system improved the accuracy of heart disease diagnoses in ultrasounds.

Researchers exposed an AI model to thousands of healthy and unhealthy ultrasound images. These decisions made by the AI model were then subject to further deep learning to enhance accuracy. 

Following the next step, which included expert diagnoses with and without AI model assistance, researchers found that the AI-based decision charts helped boost diagnosis accuracy.