KIT | KIT-Bibliothek | Impressum | Datenschutz

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

Aston, Philip J.; Mehari, Temesgen; Bosnjakovic, Alen; Harris, Peter M.; Sundar, Ashish; Williams, Steven E.; Dössel, Olaf 1; Loewe, Axel ORCID iD icon 1; Nagel, Claudia 1; Strodthoff, Nils
1 Institut für Biomedizinische Technik (IBT), Karlsruher Institut für Technologie (KIT)

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.


Verlagsausgabe §
DOI: 10.5445/IR/1000158441
Veröffentlicht am 05.05.2023
Originalveröffentlichung
DOI: 10.22489/CinC.2022.087
Scopus
Zitationen: 2
Dimensions
Zitationen: 2
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Biomedizinische Technik (IBT)
Publikationstyp Proceedingsbeitrag
Publikationsjahr 2022
Sprache Englisch
Identifikator ISBN: 979-83-503-0097-0
ISSN: 2325-887X
KITopen-ID: 1000158441
Erschienen in 2022 Computing in Cardiology Conference (CinC)
Veranstaltung 49th Computing in Cardiology (CinC 2022), Tampere, Finnland, 04.09.2022 – 07.09.2022
Verlag Institute of Electrical and Electronics Engineers (IEEE)
Serie Computing in Cardiology Conference (CinC)
Nachgewiesen in Dimensions
Scopus
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page