Prediction of Cardiac Arrest using LSTM in Hospitalized Patients

Title : Prediction of Cardiac Arrest using LSTM in Hospitalized Patients
Published in : The 4th International Conference on Interdisciplinary research on Computer science, Psychology, and Education (ICICPE’ 2020)
Author : Geonil Yun, Minsu Chae, Hyo-Wook Gil, Nam-Jun Cho, HwaMin Lee
Corresponding author : HwaMin Lee
Location : RAMADA PLAZA JEJU Hotel, Jeju Island, Korea and Online

It is necessary to detect unexpected cardiac arrest early in the general ward of the hospital. However, the conventional Track and Trigger System (TTS) has low sensitivity and a high false alarm rate, which may not guarantee the safety of heart attack patients. In addition, false alarms can lead to a waste of movement by medical staff. This can be life-threatening as other critically ill patients lose access to treatment. Therefore, in this paper, we propose an LSTM deep learning network model that predicts heart attack within 24 hours with high sensitivity and low false alarm rate using patient vital signs.

This research supported by the MSIT (Ministry of Science and ICT), Korea, under the ITRC (Information Technology Research Center) support program(IITP-2020-2015-0-00403) supervised by the IITP (Institute for Information & communications Technology Planning & Evaluation) and X-mind Corps program of National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT (No. 2019H1D8A1105622).