DeepWave: Cardiac Arrhythmia Diagnosis Using a Deep Learning ECG System
Cardiovascular diseases are the leading cause of deaths worldwide. The electrocardiogram (ECG) is commonly used to measure the electrical activity of the heart. Manual interpretation of ECG signals is often prone to errors. DeepWave presents an efficient method for interpreting ECG signals to detect cardiac arrhythmias using a recurrent neural network. DeepWave predicts arrhythmias with 90% accuracy and identifies related arrhythmia classes with similar morphologies.
Shayan Mahmood is a Grade 7 gifted student at Mountsfield Public School. This is his first CWSF and has been involved in regional science fairs since grade 4. At the Thames Valley Science and Engineering fair he won a gold medal along with 3 special awards including the IEEE award, Research Western Imagination Award and the Beckett Smith Memorial Bit-By-Bit Camp Technology Award. He loves applying STEM in creative ways to solve challenging problems. He plays a variety of sports such as soccer and basketball, loves to read books and enjoys solving math problems. He has also won many awards in math competitions run by Brock and Waterloo University. Shayan created DeepWave, a novel early diagnostic tool for cardiovascular diseases using an intelligent ECG system. This includes the development of a Recurrent Neural Network, along with a segmentation/feature extraction algorithm, which can accurately classify ECG signals into a total of 11 types of cardiac arrhythmias. He would advice other students to never ignore ideas that seem too difficult, with hard work and dedication, anything can be achieved.