Home / Health / A new tool based on AI can "detect heart failure in one beat"

A new tool based on AI can "detect heart failure in one beat"



A new tool based on AI can "detect heart failure from a single heartbeat" and is 100% accurate, say scientists

  • Scientists have fed electrocardiograms composed of 490,000 heartbeats
  • The technology was then exposed to a series of "five-minute ECG extracts".
  • Scientists hope their tool will one day help doctors diagnose HF sooner

A new tool based on AI could detect heart failure from a single heartbeat, according to research.

Scientists "fed" the system with electrocardiograms (ECGs) representing more than 490,000 heartbeats.

The technology was then exposed to a series of "five-minute ECG extracts" extracted from 24-hour records.

The results showed that the convolutional neuronal network, as it is called, was 100% accurate for screening patients with heart failure.

An AI model could detect heart failure from a heartbeat with 100% accuracy (stock)

An AI model could detect heart failure from a heartbeat with 100% accuracy (stock)

The University of Surrey team hopes its tool will one day help doctors diagnose heart failure faster, "for the benefit of patients and pressure on NHS resources".

Heart failure occurs when the muscles of the organ are too weak or too stiff to allow the blood to circulate effectively throughout the body.

This may be due to high blood pressure or narrowing of the arteries. Drinking too much alcohol can also be the cause, according to the NHS.

The condition affects about 26 million people worldwide to some extent, according to the European Society of Cardiology.

In the most severe cases, up to 40% of patients die from it, the researchers wrote.

This is also one of the leading causes of hospitalization among the elderly, the team added.

With life expectancy on the rise, the team has been working to develop a more accurate way to quickly diagnose heart failure.

Existing methods examine heart rate variability (HRV), which describes inconsistencies in the space between consecutive heartbeats.

However, one can usually only diagnose HF after examining a person's HRV for about 24 hours.

To overcome this problem, researchers led by Dr. Sebastiano Massaro focused on ECG signals rather than HRV.

They collected "long-term ECG recordings" of 15 patients with severe heart failure taken from the database on congestive heart failure BIDMC.

The "control group" consisted of 18 healthy ECGs from the normal sinus rhythm database.

Each participant had about 20 hours of ECG recordings, the researchers wrote in the journal Biomedical Signal Processing and Control.

Dr. Massaro said, "Our model has provided 100% accuracy. By checking only one heartbeat, we can detect whether a person has heart failure or not. & # 39;

He added that their tool was one of the first to be able to identify the morphological characteristics of the ECG related to the severity of heart failure.

Dr. Leandro Pecchia, President of the European Alliance for Medical and Biological Engineering, said that it offers "major progress" in the detection of heart failure.

He added: "Allowing clinical practitioners to access an accurate IC detection tool can have a significant impact on society."

Dr. Pecchia said that patients could "benefit from early and more effective diagnosis" and that the tool could alleviate "pressures on NHS resources".

HOW LEARNS ARTIFICIAL INTELLIGENCE?

AI systems rely on artificial neural networks (ANNs), which attempt to simulate how the brain works to learn.

ANNs can be trained in the recognition of information features – including speech, textual data, or visual images – and are the basis of many AI developments in recent years.

Conventional artificial intelligence uses inputs to "teach" an algorithm on a particular subject by providing it with huge amounts of information.

AI systems rely on artificial neural networks (ANNs), which attempt to simulate how the brain works to learn. ANNs can be trained to recognize patterns of information, including words, textual data, or images.

AI systems rely on artificial neural networks (ANNs), which attempt to simulate how the brain works to learn. ANNs can be trained to recognize patterns of information, including words, textual data, or images.

Practical applications include Google's language translation services, Facebook's face recognition software and Snapchat's live image editing filters.

The process of entering this data can be time consuming and limited to one type of knowledge.

A new generation of RNA called Adversarial Neural Networks opposes the spirit of two AI bots, allowing them to learn from each other.

This approach is designed to accelerate the learning process and refine the result created by artificial intelligence systems.


Source link