Show simple item record

dc.contributor.authorGaudilliere, Pierre Louis
dc.contributor.authorSigurthorsdottir, Halla
dc.contributor.authorAguet, Clémentine
dc.contributor.authorVan Zaen, Jérôme
dc.contributor.authorLemay, Mathieu
dc.contributor.authorDelgado-Gonzalo, Ricard
dc.date.accessioned2022-02-09T11:11:20Z
dc.date.available2022-02-09T11:11:20Z
dc.date.issued2021-09-13
dc.identifier.citationCinC 2021, Brno (CZ)
dc.identifier.urihttps://yoda.csem.ch/handle/20.500.12839/527
dc.description.abstractECG heartbeat classi?cation plays a vital role in diagnosis of cardiac arrhythmia. The goal of the Physionet/CinC 2021 challenge was to accurately classify clinical diagnosis based on 12, 6, 4, 3 or 2-lead ECG recordings in order to aid doctors in the diagnoses of different heart conditions. Transformers have had great success in the ?eld of natural language processing in the past years. Our team, CinCSEM, proposes to draw the parallel between text and periodic time series signals by viewing the repeated period as “words” and the whole signal as a sequence of such words. In this way, the attention mechanisms of the transformers can be applied to periodic time series signals. In our implementation, we follow the Transformer Encoder architecture, which combines several encoder layers followed by a dense layer with linear or sigmoid activation for generative pre-training or classi?cation, respectively. The use case presented here is multi-label classi?cation of heartbeat abnormalities of ECG recordings shared by the challenge. Our best entry, not exceeding the challenge’s hardware limitations, achieved a score of 0.12, 0.07, 0.10, 0.10 and 0.07 on 12lead, 6-lead, 4-lead, 3-lead and 2-lead test set respectively. Unfortunately, our team was unable to be ranked because of a missing pre-print.
dc.language.isoen
dc.titleGenerative Pre-Trained Transformer for Cardiac Abnormality Detection
dc.typeProceedings Article
dc.type.csemdivisionsDiv-E
dc.type.csemresearchareasData & AI
dc.type.csemresearchareasIoT & Vision
dc.type.csemresearchareasDigital Health
dc.identifier.urlhttp://arxiv.org/abs/2110.04071


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

  • Research Publications
    The “Research Publications” collection provides bibliographic information for scientific papers including conference proceedings and presentations.

Show simple item record