Convolutional Neural Network and Rule-Based Algorithms for Classifying 12-lead ECGs
Bjørn-Jostein Singstad, Christian Tronstad
Abstract
The objective of this study was to classify 27 cardiac abnormalities based on a data set of 43101 ECG recordings. A hybrid model combining a rule-based algorithm with different deep learning architectures was developed.
Topics & Concepts
Computer scienceConvolutional neural networkArtificial intelligenceLead (geology)AlgorithmPattern recognition (psychology)Machine learningGeologyGeomorphologyECG Monitoring and AnalysisFault Detection and Control SystemsEEG and Brain-Computer Interfaces