Prediction Model of Mortality with Respiratory Rate, Oxygen Saturation and Heart Rate using Logistic Regression

Alfi Zahra Hafizhah, Sinung Suakanto, Riska Yanu Fa'rifah, Edi Nuryatno

Research output: Chapter in Book/Conference paperConference paperpeer-review

Abstract

In the health context, sometimes we want to do an early warning of clinical deteriations. Because there are several incidents where people suddenly die without any noticeable symptoms. Or suddenly experience a drop without any obvious initial symptoms. Therefore, it is necessary to develop an initial study that can help to make predictions based on vital signs. But not all vital signs can be easily measured or in other words require the person to go to the hospital with complete equipment. It would be easier if people could make early warning predictions based on simple vital sign parameters that are respiratory rate, oxygen rate and heart rate. Where this vital sign parameter can be measured easily with existing tools without having to go to the hospital. This study aims to build a logistic regression model for predicting the mortality using oxygen saturation, respiratory rate and heart rate as parameters. Logistic regression is used because of the suitability of the model's advantages with the data, and the model evaluation uses F1- Macro. This study also uses Synthetic Minority Over-sampling technique and categorizing values of the variables to get a better model result. Training accuracy is 57%, while the evaluation accuracy is 55%. Although the accuracy is not yet good, this idea can be the basis for further development in developing early warning of clinical deterotions with limited parameters and can be done with measuring tools that are easily obtained without having to go to the hospital.
Original languageEnglish
Title of host publication2022 International Conference Advancement in Data Science, E-learning and Information Systems (ICADEIS)
PublisherIEEE, Institute of Electrical and Electronics Engineers
ISBN (Electronic)9781665463874
ISBN (Print)9781665463874
DOIs
Publication statusPublished - 2022
Event2022 International Conference Advancement in Data Science, E-learning and Information Systems - Nişantaşı University & Telkom University, Istanbul, Türkiye
Duration: 23 Nov 202224 Nov 2022
https://easychair.org/cfp/ICADEIS2022

Publication series

NameProceedings - International Conference Advancement in Data Science, E-Learning and Information Systems, ICADEIS 2022

Conference

Conference2022 International Conference Advancement in Data Science, E-learning and Information Systems
Abbreviated titleICADEIS
Country/TerritoryTürkiye
CityIstanbul
Period23/11/2224/11/22
Internet address

Fingerprint

Dive into the research topics of 'Prediction Model of Mortality with Respiratory Rate, Oxygen Saturation and Heart Rate using Logistic Regression'. Together they form a unique fingerprint.

Cite this