Abstract:

BACKGROUND: Original experiment record of cardiac electrophysiology is significant for further research. However, the original record is curve graphics. It is difficult to repeat this experiment because of the special cases or the restrictions of the experimental equipment and the cost of experiment. Therefore, it is necessary to convert the curve graphics to numerical value as a method of processing the original data in the medical signal area.
OBJECTIVE: To convert various types of electrophysiology image into numerical value by the means of MATLAB for further analysis, comparison and research store.
METHODS: Based on characteristic analysis of cardiac electrophysiology images, the cardiac electrophysiology images were subjected to numerical value, including signal extraction, noise extraction, and error correction. The last numerical values were converted matching to the actual size of the image. This algorithm was used for various types of heart electrophysiology image to prove its feasibility. Finally, the image of numerical value was compared with the original to check its accuracy.
RESULTS AND CONCLUSION: Through a comparison with the original image, the result of this algorithm could mostly reflect the original information and had a high degree of accuracy. Almost every kind of electrophysiology image could be converted using this algorithm. This algorithm is a powerful tool in image conversion. Using this algorithm, electrophysiology image can be converted to numerical value accurately and stored for further research, which is an important preparation especially for the study of heart diseases.