What is it about?
Cardiotocography (CTG) is an important procedure to monitor fetal health, usually performed in the third trimester of pregnancy. We have trained different machine-learning algorithms to classify the CTG signals into three different classes: Normal, Suspicious, and Pathological. The training of the models was done on the CTG dataset and upsampled dataset as well and then we compared the performances of the algorithms on the basis of recall, precision, f1 score, and overall accuracy.
Featured Image
Photo by Alexander Grey on Unsplash
Why is it important?
This experimental study showcases how different machine learning models perform on unbalanced and balanced forms of the same dataset for classifying the CTG records in multiple classes. It was found that when the class imbalance issue was removed from the dataset the models performed pretty well compared to the unbalanced one.
Perspectives
I wrote this paper under the guidance of my project supervisor, who is also the co-author of this paper. Data Science and Machine Learning are fields of study that can be combined with other domains to explore a lot of untapped regions to solve various problems. In this case, we used the basics to perform a task with a machine which is usually done by doctors (humans). For me on a personal level, it was very enlightening as to what we can achieve through data science and machine learning, even though this project was just the beginning step of a long journey.
Srishti Sinha
National Institute of Technology Warangal
Read the Original
This page is a summary of: Application and Evaluation of Machine Learning Algorithms in Classifying Cardiotocography (CTG) Signals, August 2023, Bentham Science Publishers,
DOI: 10.2174/9789815079210123010010.
You can read the full text:
Contributors
The following have contributed to this page
