Number of the records: 1
Classification of ECG Using Ensemble of Residual CNNs with Attention Mechanism
- 1.
SYSNO ASEP 0555163 Document Type C - Proceedings Paper (int. conf.) R&D Document Type Conference Paper Title Classification of ECG Using Ensemble of Residual CNNs with Attention Mechanism Author(s) Nejedlý, Petr (UPT-D) RID, SAI
Ivora, Adam (UPT-D)
Smíšek, Radovan (UPT-D) RID, ORCID, SAI
Viščor, Ivo (UPT-D) RID, ORCID, SAI
Koščová, Zuzana (UPT-D)
Jurák, Pavel (UPT-D) RID, ORCID, SAI
Plešinger, Filip (UPT-D) RID, ORCID, SAINumber of authors 7 Article number 14 Source Title 2021 Computing in Cardiology (CinC), 48. - New York : IEEE, 2021 - ISSN 2325-8861 - ISBN 978-166547916-5 Number of pages 4 s. Publication form Online - E Action Computing in Cardiology 2021 /48./ Event date 12.09.2021 - 15.09.2021 VEvent location Brno Country CZ - Czech Republic Event type WRD Language eng - English Country US - United States Keywords Residual CNNs ; Classification of ECG ; Physionet Challenge Subject RIV FS - Medical Facilities ; Equipment OECD category Medical engineering R&D Projects FW01010305 GA TA ČR - Technology Agency of the Czech Republic (TA ČR) Institutional support UPT-D - RVO:68081731 EID SCOPUS 85124723451 DOI 10.23919/CinC53138.2021.9662723 Annotation This paper introduces a winning solution (team ISIBrno-AIMT) to the PhysioNet Challenge 2021. The method is based on the ResNet deep neural network architecture with a multi-head attention mechanism for ECG classification into 26 independent groups. The model is optimized using a mixture of loss functions, i.e., binary cross-entropy, custom challenge score loss function, and sparsity loss function. Probability thresholds for each classification class are estimated using the evolutionary optimization method. The final model consists of three submodels forming a majority voting classification ensemble. The proposed method classifies ECGs with a variable number of leads, e.g., 12-lead, 6-lead, 4-lead, 3-lead, and 2-lead. The algorithm was validated and tested on the external hidden datasets (CPSC, G12EC, undisclosed set, UMich), achieving a challenge score 0.58 for all tested lead configurations. The total training time was approximately 27 hours, i.e., 9 hours per model. The presented solution was ranked first across all 39 teams in all categories. Workplace Institute of Scientific Instruments Contact Martina Šillerová, sillerova@ISIBrno.Cz, Tel.: 541 514 178 Year of Publishing 2022 Electronic address https://ieeexplore.ieee.org/document/9662723
Number of the records: 1