Development Of A System For Classification Of Cardiovascular Pathologies In Electrocardiographic Signals (Ecg) Applying Artificial Intelligence And Cloud Computing

In this project, the development of a system for the classification of cardiac pathologies in electrocardiographic signals (ECG) is proposed through the implementation of Artificial Intelligence (AI) based on Machine Learning under the Python programming language on Linux. The main idea of artificial intelligence is to develop methods and algorithms that allow computers to behave intelligently. In the first place, the electrocardiographic signals of healthy patients and patients with cardiovascular problems are acquired through the database called "Physiobank" such as valve disease, bundle branch block, ventricular hypertrophy and cardiac dysrhythmia, in which techniques must be applied treatment and processing of signals such as Wavelet, FFT, entropy and energy for the subsequent search for characteristics or patterns that demonstrate a difference in the signals, therefore, obtain a correct classification by applying Machine Learning techniques with respect to the heart diseases present and provide greater ease in terms of diagnosis by the specialist, based on the above, the best technique in terms of results and optimization is selected. Second, once the characterization and classification of the ECG waves is completed, Cloud Computing strategies are applied to manage the data, store and process it under the same server and deliver results online.

The ECG signals are worked in .mat format, these are pre-processed and processed using signal treatment techniques, techniques such as Support Vector Machine, Naïve Bayes and Decision Trees are implemented, on the other hand, the selection of the technique to be used as classification is based on the best classification percentage produced by the model generated by each of the aforementioned techniques, once the best technique has been selected, the signal classifier system is terminated and a web application is created using the Use of a micro-framework called Flask which can be implemented in Python, as an advantage code can be executed under said programming language allowing you to create your own web page and be able to be compiled in the cloud through the Cloud Pythonanywhere service.