Establishment of a Real-Time Medical Resource Information Platform

Overview

Experiencing pandemics and other crises has heightened the need to secure the golden hour in emergencies and connect patients to hospitals capable of timely treatment. Consequently, there is a growing demand to improve hospital systems and enhance the real-time accuracy of information regarding healthcare institutions' emergency patient capacity.

In particular, commercialized AI machine learning solutions lack algorithms specialized for medical analysis. The need to develop an AI-based predictive model for initial emergency patient admission to prevent emergency room overcrowding has also been raised.

Project description

To manage detailed Hospital Information System (HIS) data for four critical conditions including cardiac arrest and transmit it in real time to the Central Emergency Medical Response System (EMRIS), a cloud-based real-time medical resource information platform has been established. This platform clearly displays essential information crucial for timely patient transport during emergencies—such as 'bed availability, essential medical equipment operational status, and capacity for major critical illnesses'—key information for timely patient transfer during emergencies—along with clearly indicating the actual availability of medical personnel and providing this information intuitively to hospital staff.

• Real-time Transmission of Medical Information
  • Improved the hospital system to transmit ward and equipment information to the Emergency Medical Integrated Situation Board in real-time within one minute.
• Detailed bed/equipment information
  • Managed detailed information critical for timely patient transfer during emergencies—such as bed availability, operational status of essential medical equipment, and capacity for major critical illnesses—alongside the actual availability of medical personnel.
• Enhanced Critical Condition Acceptance Capabilities
  • Advanced functionality to confirm and manage acceptance for four major critical conditions: cardiac arrest, subarachnoid hemorrhage, obstetric emergencies, and acute aortic diseases
• Integrated Dashboard Development
  • Provide stakeholders with an intuitive overview of the actual availability of beds, equipment, and medical personnel essential for treating emergency patients, including operating room vacancies and equipment malfunctions.
• Development of Initial Emergency Patient Admission Prediction Model
  • Develop a prediction model for initial emergency patient admission by leveraging AI- integrated monitoring to learn normal patterns based on emergency room overcrowding, detect performance and service anomalies, identify overcrowding risk indices, and analyze root causes.

Key technological features

• Real-time Medical Resource Information Collection and Integrated Dashboard Implementation
  • Implementation of an integrated dashboard through AI analysis of medical resource information, including: inpatient ward operational status, operating room/delivery room operational status, equipment (CT, MRI, ventilators, etc.) operational status, critical illness admission status, and medical personnel operational status
• Hospital Information System Enhancement and Data Integration with the Emergency Medical Comprehensive Situation Board

Expected outcomes

This service will integrate and manage scattered medical resource information within healthcare institutions, enabling timely patient transfer to facilities capable of providing treatment during disasters and emergencies. It will significantly contribute to securing the golden hour for patients through appropriate initial response.

It will also reduce unnecessary economic costs associated with 119 ambulance dispatch, patient transport, and comprehensive emergency medical situation assessment. Furthermore, the AI-based emergency patient admission prediction model is expected to be utilized for efficient emergency medical resource allocation during outbreaks of Class 1 statutory infectious diseases such as Ebola virus and SARS.

Launch year: 2024

Attribution

Case study courtesy of the National Information Society Agency, Republic of Korea, Apolitical's Content Partner. Sourced from Use Cases of Public AI Service, Vol.1,  https://eng.nia.or.kr/site/nia_eng/ex/bbs/View.do?cbIdx=31975&bcIdx=28997&parentSeq=29029