Dynamic behavior of RR and QT intervals in the electrocardiogram
Kokkonen, Jimi (2023)
Kokkonen, Jimi
2023
Teknis-luonnontieteellinen DI-ohjelma - Master's Programme in Science and Engineering
Tekniikan ja luonnontieteiden tiedekunta - Faculty of Engineering and Natural Sciences
This publication is copyrighted. Only for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2023-11-07
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202311019328
https://urn.fi/URN:NBN:fi:tuni-202311019328
Tiivistelmä
The electrocardiogram (ECG) is a plot of the heart's electrical activity as a function of time and a common method for identifying health issues. The ECG consists of a series of five waves and a number of intervals, each representing a certain period of the electrical signal traveling through the heart during a single heartbeat. By measuring these intervals, a number of health issues affecting both the heart and the rest of the body can be diagnosed.
This study focuses on the RR and QT intervals of the ECG. QT intervals has been observed to have a relationship with the heart rate and therefore the RR interval in a way that is highly subject-specific. Accurate determination of the QT interval is important as it can be used to diagnose various arrhythmias, such as long QT syndrome indicated by an abnormally long QT interval. To improve the methods used to calculate heart rate-corrected QT intervals used for diagnosis, it is useful to gain a deeper understanding of the relationship between RR and QT intervals.
In this thesis, the relationship of the RR and QT intervals is examined in two datasets of ECG recordings. One dataset contains a large set of 24-hour ambulatory recordings associated with a variety of daily activities such as sleep, work and exercise. The other dataset is a smaller, more limited set of recordings roughly an hour long. The cross-correlations of RR and QT intervals is examined both at the level of individual heartbeats and collectively throughout an entire dataset. In addition, the autocorrelations, that is, the correlations of a time series with itself over a certain time range, of the individual RR and QT time series are examined and compared with detrended fluctuation analysis (DFA) as well as dynamical detrended fluctuation analysis (DDFA). The analysis was done using the Python programming language and related libraries.
It is observed that the autocorrelations within RR and QT intervals behave differently as a function of the scale. There is also a difference in the autocorrelations between datasets, likely as a result of the different conditions in which they were recorded. The results also suggest that the impact of RR intervals on each successive QT interval is stronger than that of QT intervals on each successive RR interval. In addition, the cross-correlation between RR and QT intervals appears to decline with age. The RR time series are also observed to have distinct recurring events with a strong impact on the RR-QT interval, the study of which may provide valuable new information in further studies.
This study focuses on the RR and QT intervals of the ECG. QT intervals has been observed to have a relationship with the heart rate and therefore the RR interval in a way that is highly subject-specific. Accurate determination of the QT interval is important as it can be used to diagnose various arrhythmias, such as long QT syndrome indicated by an abnormally long QT interval. To improve the methods used to calculate heart rate-corrected QT intervals used for diagnosis, it is useful to gain a deeper understanding of the relationship between RR and QT intervals.
In this thesis, the relationship of the RR and QT intervals is examined in two datasets of ECG recordings. One dataset contains a large set of 24-hour ambulatory recordings associated with a variety of daily activities such as sleep, work and exercise. The other dataset is a smaller, more limited set of recordings roughly an hour long. The cross-correlations of RR and QT intervals is examined both at the level of individual heartbeats and collectively throughout an entire dataset. In addition, the autocorrelations, that is, the correlations of a time series with itself over a certain time range, of the individual RR and QT time series are examined and compared with detrended fluctuation analysis (DFA) as well as dynamical detrended fluctuation analysis (DDFA). The analysis was done using the Python programming language and related libraries.
It is observed that the autocorrelations within RR and QT intervals behave differently as a function of the scale. There is also a difference in the autocorrelations between datasets, likely as a result of the different conditions in which they were recorded. The results also suggest that the impact of RR intervals on each successive QT interval is stronger than that of QT intervals on each successive RR interval. In addition, the cross-correlation between RR and QT intervals appears to decline with age. The RR time series are also observed to have distinct recurring events with a strong impact on the RR-QT interval, the study of which may provide valuable new information in further studies.