Please use this identifier to cite or link to this item:

http://hdl.handle.net/10609/133088
Title: Extraction of dynamical patterns from fluorescence microscopy images using recurrent neural networks
Author: Ospina Mesa, Andrés
Tutor: Benítez Iglesias, Raúl
Keywords: deep learning
anomaly detection
autoencoder
signals processing
artificial intelligence
Issue Date: Jun-2021
Publisher: Universitat Oberta de Catalunya (UOC)
Abstract: Heart beating it's what keeps human living, the correct function of it, among other things, determinate the life quality, longevity and diseases appearing, what is known is that arrhythmia it's the most common disease in the cardiac system, what it's not so common is that can be detected earlier, it can exist and even not be visible in electrocardiogram, that's because the heart contractions occur at cellular level, then at tissue level and finally at muscular level, if the arrhythmia test it's done at muscular level, that means that can be observed also at cellular and tissue level, the objective of this work it is using Recurrent Neural Networks(RNN), more precisely, Autoencoders + LSTM, to identify anomalies, which can be arrhythmia or other diseases, based on electrophysiological signals at cellular level, extracted from patients' cardiac tissue, creating interactive data visualization such as dashboard in Hypertext Markup Language(HTML) format.
Language: English
URI: http://hdl.handle.net/10609/133088
Appears in Collections:Bachelor thesis, research projects, etc.

Share:
Export:
Files in This Item:
File Description SizeFormat 
aospinameTFM0621memory.pdfTFM Memory1.59 MBAdobe PDFView/Open

This item is licensed under a Creative Commons License Creative Commons