Multi-Class ECG Feature Importance Rankings: Cardiologists vs. Algorithms

Philip J. Aston; Temesgen Mehari; Alen Bosnjakovic; Peter M. Harris; Ashish Sundar; Steven E. Williams; Olaf Dossel; Axel Loewe; Claudia Nagel; Nils Strodthoff

doi:10.22489/cinc.2022.087

Back

Multi-Class ECG Feature Importance Rankings: Cardiologists vs. Algorithms

Conference proceeding

Open access

Multi-Class ECG Feature Importance Rankings: Cardiologists vs. Algorithms

Philip J. Aston, Temesgen Mehari, Alen Bosnjakovic, Peter M. Harris, Ashish Sundar, Steven E. Williams, Olaf Dossel, Axel Loewe, Claudia Nagel and Nils Strodthoff

Computing in Cardiology (CinC), 2012, Vol.49

2022

DOI: https://doi.org/10.22489/cinc.2022.087

Abstract

Cardiologists have been using electrocardiogram features to diagnose a wide variety of heart conditions for many decades. There are also numerous algorithms that rank feature importance for a particular classification task. However, different algorithms often give quite different feature rankings. Therefore, we compared the feature importance rankings obtained by various algorithms with the features that cardiologists use for diagnosis.

Files and links (1)

url

https://doi.org/10.22489/cinc.2022.087View

Published (Version of record) Open

Metrics

18 Record Views

Details

Title: Multi-Class ECG Feature Importance Rankings: Cardiologists vs. Algorithms
Creators: Philip J. Aston - University of Surrey
Temesgen Mehari - Heinrich Hertz Institute
Alen Bosnjakovic - Institute of Metrology of Bosnia and Herzegovina
Peter M. Harris - University of Surrey
Ashish Sundar - University of Surrey
Steven E. Williams - University of Edinburgh
Olaf Dossel - Karlsruhe Institute of Technology
Axel Loewe - Karlsruhe Institute of Technology
Claudia Nagel - Karlsruhe Institute of Technology
Nils Strodthoff - Carl von Ossietzky Universität Oldenburg
Publication Details: Computing in Cardiology (CinC), 2012, Vol.49
Publisher: IEEE Computer Society
Date published: 2022
Grant note: This project 18HLT07 MedalCare has received funding from the EMPIR programme co-financed by the Participating States and from the European Union’s Horizon 2020 research and innovation programme. The authors acknowledge the support of the British Heart Foundation Centre for Research Excellence Award III (RE/18/5/34216). This research is supported by the British Heart Foundation (RE/18/5/34216). This research is part of the British Heart Foundation Centre for Research Excellence at The University of Edinburgh (RE/18/5/34216). SEW is supported by the British Heart Foundation (FS/20/26/34952).
Identifiers: 99822525002346
Academic Unit: School of Maths and Physics
Language: English
Resource Type: Conference proceeding

Multi-Class ECG Feature Importance Rankings: Cardiologists vs. Algorithms

Abstract

Files and links (1)

Metrics

Details

Usage Policy