Biomedical and Mechanical Engineering Journal
Vol. 2 No. 2 (2022): BIOMEJ

Artificial Neural Network Application for Sepsis Prediction: A Preliminary Study

A K Faizin (Unknown)
R D Issafira (Unknown)
W Saputro (Unknown)
L Edahwati (Unknown)
W D Lestari (Unknown)
T P Sari (Unknown)
N Adyono (Unknown)



Article Info

Publish Date
17 Nov 2022

Abstract

Abstract. Sepsis is one of the leading causes of mortality in hospitalized patients. It is very difficult to find the symptoms of sepsis because of their similarity to the symptoms of other diseases. This paper aims to deliver an artificial neural network implementation in medical decisions support. This study tries to predict sepsis and healthy patient based on vital signs such as heart rate, systolic blood pressure, and diastolic blood pressure taken from the MIMIC-III clinical database. There were several extraction processes applied to vital sign signals such as using the statistical tools, discrete wavelet transforms, and Hilbert-Huang Transform. The ANN algorithm predicts the sepsis patient with 96.7% of accuracy. However, based on the medical requirement for artificial intelligent implementation, this result does not satisfy the requirement as the false positive error is 2.9%.

Copyrights © 2022






Journal Info

Abbrev

biomej

Publisher

Subject

Automotive Engineering Chemical Engineering, Chemistry & Bioengineering Civil Engineering, Building, Construction & Architecture Electrical & Electronics Engineering Energy Engineering Industrial & Manufacturing Engineering Materials Science & Nanotechnology Mechanical Engineering Transportation

Description

The Biomej Journal is published by the Mechanical Engineering Study Program, Faculty of Engineering, East Java "Veteran" University, Surabaya-Indonesia. Biomej is an open-access peer review journal that mediates the dissemination of academics, researchers, and practitioners in the field of ...