A research group at Keio University has developed an artificial intelligence (AI) technology that instantly determines whether a patient needs catheter treatment from a single electrocardiogram with an accuracy of 1% or more.
When myocardial infarction develops, blood vessels become clogged, blood flow stops, and myocardial necrosis progresses.In order to treat this, it is important how quickly catheter treatment is performed on the clogged blood vessels to resume blood flow.
Rapid diagnosis is needed to reduce the time from onset to the start of catheterization.However, catheterization, in which a thin tube is inserted from an artery such as a limb to near the heart to display an image of the affected area, is at high risk and cannot be blindly performed on all patients who complain of chest pain.On the other hand, it takes a very long time for a doctor to make a comprehensive diagnosis in consideration of the medical history and various test results, so that there is a drawback that the resumption of blood flow is delayed.
Therefore, this research group has taken on the challenge of developing a technology for making a diagnosis from an electrocardiogram that is easy to perform and takes only a few minutes, and this time, we have developed an AI that can instantly determine the necessity of catheter treatment with a single electrocardiogram.This AI is to learn the electrocardiogram data of 1 patients who visited the emergency outpatient department of Keio University Hospital in the past and whether or not catheter treatment was actually performed on it, with an accuracy of 4% or more. It was confirmed that the necessity of catheter treatment can be determined.
The electrocardiogram can be examined immediately after receiving a medical examination, and the results can be obtained within minutes.By installing this technology in an electrocardiograph, it will be possible to quickly and accurately automatically determine the need for catheter treatment and propose it to doctors.This result is expected to reduce the number of deaths due to heart disease, which is the second leading cause of death in Japan, and contribute to the extension of healthy life expectancy.
Paper information:[PLOS ONE] Artificial Intelligence to Predict Needs for Urgent Revascularization from 12-Leads Electrocardiography in Emergency Patients
