[ad_1]
MediPlex Sejong Hospital Department of Emergency Medicine Kwon Jun Myung checks the condition of hospitalized patients by looking at the real-time monitoring system risk of cardiac arrest. The Sejong Hospital
DEWS, a real-time monitoring system of risk indications of patients using artificial intelligence developed at the national level, has detected the risk of cardiac arrest in hospitalized patients at least 14 hours before helping to prevent cardiac arrest. ] The director of the Department of Emergency Medicine at Sejong Hospital ) studied the feasibility of Deuces in collaboration with Lee's research team, Yea Ha
In a clinical study, Two not only showed a 50% chance of detecting cardiac arrest 14 hours ago, but also a sensitivity 24% higher than a conventional mechanical alarm system.
The results of the study were published in the latest issue of the Journal of the American Heart Association (JAHA), an international journal published by the American Heart Association.
The newest method of machine learning, Deep Learning, which is the core technology of Alpha Go, can predict the risk of cardiac arrest in patients hospitalized to at least 50%
The team developed an artificial intelligence algorithm that predicts in advance the patient's cardiac arrest based on systolic blood pressure, the pulse rate, respiratory rate and body temperature of the hospitalized patient, The performance was also confirmed. He also found that more than 50% of patients with cardiac arrest had been discovered before 14 hours, 24% more sensitive and 24% less false alarms than the previous method.
Kwon Jun Myung, director of the emergency department of Sejong Medical Center, said, "I can estimate the risk of cardiac arrest by estimating only 4 values, so that" it can be industrialized as a portable device.
The MediPLEX Sejong Hospital is currently part of the rapid response team, which should be available soon.In the second half of this year, we used Deuce as an early predictor of patient abnormalities I will ask permission to apply for medical devices
KiSu Su medical reporter [email protected]
[ad_2]
Source link
