A newly developed machine learning model uses electrocardiogram (ECG) readings to diagnose and classify heart attacks more accurately and faster than current methods.
The study, led by researchers from the University of Pittsburgh and published in Nature Medicine, suggests significant improvements in patient care can be achieved using this model.
Addressing a Major Challenge in Heart Attack Diagnosis
The first question asked when a patient with chest pain arrives at the hospital is whether they are experiencing a heart attack.
Although this seems straightforward, the answer isn’t always clear from the initial ECG, leading to additional testing that can take up to 24 hours.
This machine learning model aims to improve risk assessment, allowing for more immediate and appropriate care.
Recognizing Subtle Clues in ECGs
While certain patterns in an ECG signal a severe type of heart attack known as STEMI, nearly two-thirds of heart attacks do not exhibit these obvious ECG patterns, despite being caused by severe artery blockage.
The machine learning model helps clinicians detect subtle signs in the ECG that are often challenging to identify, thus enhancing patient classification.
Developing and Testing the Model
The model was developed using ECG data from 4,026 patients with chest pain across three hospitals in Pittsburgh.
It was then externally validated using data from an additional 3,287 patients from a different hospital system.
Researchers compared the model’s performance to three gold standards for assessing cardiac events: experienced clinician interpretation of ECG, commercial ECG algorithms, and the HEART score, a comprehensive risk assessment tool.
Surprisingly, the machine learning model outperformed all three, accurately reclassifying one in three patients as low, intermediate, or high risk.
Implementing the Model in Emergency Services
The model’s success suggests that it can assist emergency medical personnel and emergency department providers in identifying heart attacks and cases of reduced blood flow to the heart more effectively than traditional ECG analysis.
For the next phase of the research, the team is partnering with the City of Pittsburgh Bureau of Emergency Medical Services to optimize how the model will be deployed.
They are developing a cloud-based system that can analyze ECG readings from EMS and send back a risk assessment of the patient, helping guide real-time medical decisions.
If you care about heart health, please read studies about how to control your cholesterol to prevent heart attacks and strokes, and how to reverse heart failure with diet.
For more information about heart health, please see recent studies that green tea may protect your body as a vaccine, and results showing this inexpensive drug combo can protect your heart health, and prevent stroke.
The study was published in Nature Medicine. Follow us on Twitter for more articles about this topic.
Copyright © 2023 Knowridge Science Report. All rights reserved.
