Speaker
Description
The human body is, without a doubt, one of the most intriguing multi-level machines. The whole organization, from the smaller, highly complex unit, the cell, to the larger and well-coordinated building blocks that operate all functions in the body, the organs, has been the object of many studies. For my master thesis, I want to focus my work on the heart, the central organ of the circulatory system and life support for the whole body.
To aid in the analysis of the electrical signal in the heart, feature-based machine learning techniques will be integrated. All data will be taken from patients that have been followed by Santa Marta’s Hospital. To substitute the invasive method of catheterization, the focus will be on developing a model to interpret ECG data. As ECG data is inherently noisy, noise filtering methods will be explored to find an optimal and efficient way to pre-process the data. During the initial phase of this thesis, ECG signal data went through a process of noise filtering using a low-pass noise filter with a threshold of 15Hz, which gave a promising end-result.
The model created is expected to be able to classify ECG interpretable cardiopathies and to implement in real-time to aid physicians in their labor.
