Analýza a vizualizace databází se zaměřením na krevní tlak

Abstract

This thesis focuses on the analysis and evaluation of publicly available databases containing blood pressure recordings. The main objective was to develop an application enabling automated processing, analysis, and visualization of large-scale physiological data. Using the developed application, 3,602 records with a total duration of 24,249 hours were analyzed. Significant differences in data quality were observed among the databases, with those obtained under controlled conditions reaching up to 85% of valid data, while intensive care unit databases contained more artifacts and lower signal usability. Differences in physiological values were also identified, such as higher blood pressure in the CHARIS database and more stable values in the AGING database. Based on the results, the databases were evaluated and recommendations for their use in research were formulated.

Description

Delayed publication

Available after

Subject(s)

databases, blood pressure, MATLAB, PhysioNet, application, MIMIC, CHARIS, VTaC, Autonomic Aging, MGH/MF

Citation