Cardiotocography waveform analysis using image extraction technique

Abstract

Cardiotocography (CTG) is one of the common tools used to monitor mother and fetus health status. CTG displays the continuous reading of fetal heart rate (FHR) and uterine contractions (UC) in graphical form. The common parameters used to analyze the signals are baseline, variability, accelerations, and decelerations of FHR. By using these parameters, the overall impression of the CTG waveform can be determined. Failure to accurately determine an abnormal condition might lead to a severe problem for the mother or fetus. This paper describes a CTG waveform analysis tool for easier monitoring of the fetal heart rate and uterine contractions. MATLAB is used to create the Graphical User Interface (GUI) to help users in using the developed algorithm more efficiently. The developed algorithm is used to extract the data from an image file of a CTG waveform printout to a useful signal. In this work, two methods are used to determine uterine contractions periods, acceleration, and deceleration. It is found that method 1 takes an average of 0.02786 seconds, while method 2 takes an average 0.01175 seconds to process the image files. The code length of method 1 in analyzing accelerations and decelerations is 10 rows, and method 2 only requires 4 and 2 rows for analyzing accelerations and decelerations, respectively. The accuracy of both methods is similar. Hence, it is concluded that method 2 with a built-in function is a better solution.