Correlation Analysis Between Vital Signs Of Patients In Intensive Care Unit Yoğun Bakım Hastalarının Hayati Bulguları Arasındaki Korelasyon Analizi


OLCAY F. F., Duru D. G., Mühendisliği B., Biyoteknoloji M.

32nd IEEE Conference on Signal Processing and Communications Applications, SIU 2024, Mersin, Turkey, 15 - 18 May 2024 identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/siu61531.2024.10601026
  • City: Mersin
  • Country: Turkey
  • Keywords: data processing, health, heart rate, MIMIC-III clinical database, non-invasive vital signs, pulse, SpO2, time series
  • Yıldız Technical University Affiliated: Yes

Abstract

Artificial intelligence (AI) and machine learning (ML) techniques have become very important in healthcare and health data processing. The proliferation of the internet and technological advancements has led to the need to use various platforms to fulfill daily routines. As a result, there is an increase in personalized user experiences, especially in healthcare monitoring through non-invasive vital signs analysis. The widespread use of artificial intelligence and machine learning technologies in the field has increased the accuracy and accessibility of the analysis of health monitoring systems, especially by leveraging comprehensive databases such as the MIMIC-III Clinical Database (v1.4), thus eliminating the need for measurements or additional sensors for data collection. In this study, the MIMIC-III waveform database was used to examine the relationship between patients' non-invasive vital signs and the intensive care units in which they were admitted. The correlation information obtained started with the comparison of raw versions of the data and continued with mathematical transformations such as Fourier transforms to detect hidden patterns and search for significant relationships between vital signs in many dimensions, and the results are reported in this study. This correlation can be used to optimize data processing and hyperparameter settings before using machine learning and deep learning techniques in vital signs analysis.