CLINICAL APPLICATION OF A SOFTWARE DEVELOPED FOR OBTAINING ELECTROCARDIOGRAPHIC SIGNALS

Abstract

BackgroundThe electrocardiogram (ECG) is widely used in diagnosis, planning of cardiovascular therapeutic intervention and evaluation of heart diseases. The aim of this study is to present clinical application and validation of a software developed for obtaining electrocardiographic signals from printed ECGs. The relevance of the developed software is related to the importance of free tools to obtain electrocardiographic signs, especially as a method to support researches, in order to allow the computational processing, the development of tools and Methods for analysis of these signals. This research presents the software development and the applications related to its use in clinical practice.Method and ResultsThe software for obtaining electrocardiographic signals was developed in Matlab, it has a graphical interface for rapid and easy use by health professionals. General steps of image processing to extract the waveform of the electrocardiogram consist of the following steps: 1. the image is converted to grayscale; 2. the background image is removed by thresholding; 3. the waveform is obtained by converting the values of pixels in an arbitrary amplitude scale; 4. interpolation is applied to the obtained signal in order to ensure there are no gaps in the signal; 5. the signal is normalized and sampled in accordance with parameters pre-set by the user. Printed electrocardiograms must be first converted into digital images for further manipulation in the program. The validation of this method was carried out with ECG records of 40 patients. The printed ECGs were scanned and electrocardiographic signals obtained from the images. It was compared the measurements of the manual values obtained to P wave duration, QRS duration and T wave duration in the printed ECG and the values of the obtained digital ECG. It was performed the t-test for a sample mean to compare the values; obtained manually and those obtained from the developed software. The obtained results showed that the developed software is a useful and relevant tool for the reconstruction of digital ECG from printed waveforms. It was obtained a p-value greater than 0.05 (p > 0.05) and the null hypothesis was accepted, therefore the statistical analysis, comparing estimates from digital ECG and manual measurements, suggested no relevant differences in the estimate of the duration of the P wave, QRS interval and T wave.ConclusionThe developed software is an efficient method for obtaining electrocardiographic signals and it can be a useful application for clinical use, cardiovascular researches and signal processing. BackgroundThe electrocardiogram (ECG) is widely used in diagnosis, planning of cardiovascular therapeutic intervention and evaluation of heart diseases. The aim of this study is to present clinical application and validation of a software developed for obtaining electrocardiographic signals from printed ECGs. The relevance of the developed software is related to the importance of free tools to obtain electrocardiographic signs, especially as a method to support researches, in order to allow the computational processing, the development of tools and Methods for analysis of these signals. This research presents the software development and the applications related to its use in clinical practice. The electrocardiogram (ECG) is widely used in diagnosis, planning of cardiovascular therapeutic intervention and evaluation of heart diseases. The aim of this study is to present clinical application and validation of a software developed for obtaining electrocardiographic signals from printed ECGs. The relevance of the developed software is related to the importance of free tools to obtain electrocardiographic signs, especially as a method to support researches, in order to allow the computational processing, the development of tools and Methods for analysis of these signals. This research presents the software development and the applications related to its use in clinical practice. Method and ResultsThe software for obtaining electrocardiographic signals was developed in Matlab, it has a graphical interface for rapid and easy use by health professionals. General steps of image processing to extract the waveform of the electrocardiogram consist of the following steps: 1. the image is converted to grayscale; 2. the background image is removed by thresholding; 3. the waveform is obtained by converting the values of pixels in an arbitrary amplitude scale; 4. interpolation is applied to the obtained signal in order to ensure there are no gaps in the signal; 5. the signal is normalized and sampled in accordance with parameters pre-set by the user. Printed electrocardiograms must be first converted into digital images for further manipulation in the program. The validation of this method was carried out with ECG records of 40 patients. The printed ECGs were scanned and electrocardiographic signals obtained from the images. It was compared the measurements of the manual values obtained to P wave duration, QRS duration and T wave duration in the printed ECG and the values of the obtained digital ECG. It was performed the t-test for a sample mean to compare the values; obtained manually and those obtained from the developed software. The obtained results showed that the developed software is a useful and relevant tool for the reconstruction of digital ECG from printed waveforms. It was obtained a p-value greater than 0.05 (p > 0.05) and the null hypothesis was accepted, therefore the statistical analysis, comparing estimates from digital ECG and manual measurements, suggested no relevant differences in the estimate of the duration of the P wave, QRS interval and T wave. The software for obtaining electrocardiographic signals was developed in Matlab, it has a graphical interface for rapid and easy use by health professionals. General steps of image processing to extract the waveform of the electrocardiogram consist of the following steps: 1. the image is converted to grayscale; 2. the background image is removed by thresholding; 3. the waveform is obtained by converting the values of pixels in an arbitrary amplitude scale; 4. interpolation is applied to the obtained signal in order to ensure there are no gaps in the signal; 5. the signal is normalized and sampled in accordance with parameters pre-set by the user. Printed electrocardiograms must be first converted into digital images for further manipulation in the program. The validation of this method was carried out with ECG records of 40 patients. The printed ECGs were scanned and electrocardiographic signals obtained from the images. It was compared the measurements of the manual values obtained to P wave duration, QRS duration and T wave duration in the printed ECG and the values of the obtained digital ECG. It was performed the t-test for a sample mean to compare the values; obtained manually and those obtained from the developed software. The obtained results showed that the developed software is a useful and relevant tool for the reconstruction of digital ECG from printed waveforms. It was obtained a p-value greater than 0.05 (p > 0.05) and the null hypothesis was accepted, therefore the statistical analysis, comparing estimates from digital ECG and manual measurements, suggested no relevant differences in the estimate of the duration of the P wave, QRS interval and T wave. ConclusionThe developed software is an efficient method for obtaining electrocardiographic signals and it can be a useful application for clinical use, cardiovascular researches and signal processing. The developed software is an efficient method for obtaining electrocardiographic signals and it can be a useful application for clinical use, cardiovascular researches and signal processing.