Utilize este identificador para referenciar este registo:
http://hdl.handle.net/10071/28412
Autoria: | Costa, J. L. Couto, P. Rodrigues, R. |
Editor: | Hyttinen, J., Paci, M., and Koivumäki, J. |
Data: | 2022 |
Título próprio: | Multitask and transfer learning for cardiac abnormality detection in heart sounds |
Volume: | 49 |
Título e volume do livro: | 49th Computing in Cardiology Conference |
Título do evento: | 49th Computing in Cardiology Conference |
Referência bibliográfica: | Costa, J. L., Couto, P., & Rodrigues, R. (2022). Multitask and transfer learning for cardiac abnormality detection in heart sounds. In J. Hyttinen, M. Paci, & J. Koivumäki (Eds.), 49th Computing in Cardiology Conference. Computing in Cardiology. https://doi.org/10.22489/CinC.2022.193 |
ISSN: | 2325-887X |
DOI (Digital Object Identifier): | 10.22489/CinC.2022.193 |
Resumo: | We present a deep learning model for the automatic detection of murmurs and other cardiac abnormalities from the analysis of digital recordings of cardiac auscultations. This approach was developed in the context of the George B. Moody PhysioNet Challenge 2022. More precisely, we consider multi-objective neural networks, with several Transformer blocks at their core, trained to perform 3 distinct tasks simultaneously: murmur detection, outcome classification and audio signal segmentation. We also perform pre-training with the 2016’s Challenge data. We entered the challenge under the team name matLisboa. Our results on the hidden test dataset were: Murmur score (weighted accuracy): 0.735 (ranked 15th). Outcomes score (cost): 12593 (ranked 16th). |
Arbitragem científica: | yes |
Acesso: | Acesso Aberto |
Aparece nas coleções: | BRU-CRI - Comunicações a conferências internacionais |
Ficheiros deste registo:
Ficheiro | Tamanho | Formato | |
---|---|---|---|
conferenceobject_95288.pdf | 268,19 kB | Adobe PDF | Ver/Abrir |
Todos os registos no repositório estão protegidos por leis de copyright, com todos os direitos reservados.