Groundbreaking ECG analysis predicts risk of death with 85% accuracy

[Apr. 2, 2023: JD Shavit, The Brighter Side of News]

An exciting example of AI in healthcare is its application to electrocardiograms (ECGs), which are used to monitor and diagnose heart health. (CREDIT: Creative Commons)

Artificial intelligence (AI) is making inroads in the healthcare industry in various ways. From improving medical diagnostics to finding new cures for diseases, AI is revolutionizing the way medical professionals approach their work.

An exciting example of AI in healthcare is its application to electrocardiograms (ECGs), which are used to monitor and diagnose heart health. Researchers in northern Alberta, Canada, are using AI to glean more insights from ECGs and improve patient care and the healthcare system as a whole.

ECGs are a standard test in hospitals, used to check the rhythm and electrical activity of the heart. Hospital staff usually read the results, but researchers are now using AI to analyze the data and identify patients who are at higher risk of death.

The research team built and trained machine learning programs based on 1.6 million ECGs performed on 244,077 patients in northern Alberta between 2007 and 2020. The algorithm predicted the risk of death from that point for each patient, all causes, in one month, one year and five years. years with an accuracy rate of 85%.

Related stories

It sorted patients into five categories, from lowest risk to highest risk, with even more precision when demographic information (age and gender) and six standard lab blood test results (creatinine, kidney function, sodium , troponin, hemoglobin and potassium) were included.

This study is a proof of concept for using routinely collected data to improve individual care and enable the health system to learn as it goes. The researchers concluded that machine learning models can be used to convert data routinely collected in clinical practice into insights that can be used to improve point-of-care decision-making within a healthcare system. learner.

Study lead researcher Padma Kaul, a professor of medicine and co-director of the Canadian VIGOR Center, says she wanted to know if she could use new AI and machine learning methods to analyze data and identify patients. which are at higher levels. mortality risk.

(Left to right) Padma Kaul with research team members Sunil Kalmady Vasu, Nariman Sepehrvand and Weidie Sun. The team has developed a machine learning program that can accurately predict a patient’s risk of death within a month, a year and five years based on the results of routine hospital tests. . (CREDIT: University of Calgary)

In the first phase of the study, the ECG results of all patients were examined, but Kaul and his team hope to refine these models for particular subgroups of patients. They also plan to focus forecasts beyond all-cause mortality to look specifically at heart-related causes of death.

Kaul says Alberta is well positioned to analyze population-level data because so much is collected and archived in the publicly funded health care system. According to Kaul, there’s a lot of pressure to see how AI can improve healthcare delivery, and for Albertans, it’s really a demonstration of their data at work.

AUROC model performance comparison for DL ​​and XGB models with ECG traces and measurements. (CREDIT: NPJ Digital Medicine)

The advantage of using a high-powered computer is that, unlike humans, it can see patterns in a multitude of data points at once. Kaul says they want to take the data generated by the healthcare system, convert it into knowledge, and feed it back into the system so they can improve care and outcomes. This is the definition of a learning health system.

ECGs are usually ordered if a patient has high blood pressure or symptoms of heart disease, such as chest pain, shortness of breath, or irregular heartbeat. Thanks to AI, healthcare professionals will be able to identify patients who are at higher risk of mortality and take preventive measures to reduce this risk.

Model performance in diagnostic and sex-based subpopulations. (CREDIT: NPJ Digital Medicine)

Additionally, AI will enable better use of healthcare resources by identifying those most at risk of mortality and in need of immediate attention.

The study conducted by Kaul and his team is only the beginning of the use of AI in ECG analysis.

As AI technology continues to develop and improve, the possibilities for use in healthcare are endless. AI will enable healthcare professionals to provide more personalized care to patients, resulting in better health outcomes and a more efficient healthcare system.

For more science and technology stories, check out our New Discoveries section on The bright side of the news.

Note: The documents provided above by The bright side of the news. Content may be edited for style and length.

Do you like these kind of wellness stories? Get the Brighter Side of News Newsletter.

Leave a Comment