The paper benchmarks CNNs and vision transformers for congenital heart disease classification on the ZCHSound and DICOM datasets, achieving 73.9% and 80.72% accuracy respectively.
The heart sound dataset consists of 941 participants and 941 audio recordings, each approximately 20 seconds long, totaling over 5 hours in duration
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Congenital Heart Disease recognition using Deep Learning/Transformer models
The paper benchmarks CNNs and vision transformers for congenital heart disease classification on the ZCHSound and DICOM datasets, achieving 73.9% and 80.72% accuracy respectively.