A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
Classification of Heart Sounds Using Convolutional Neural Network (2020)


Li, Fan; Tang, Hong; Shang, Shang; Mathiak, Klaus; Cong, Fengyu (2020). Classification of Heart Sounds Using Convolutional Neural Network. Applied Sciences, 10 (11), 3956. DOI: 10.3390/app10113956


JYU-tekijät tai -toimittajat


Julkaisun tiedot

Julkaisun kaikki tekijät tai toimittajat: Li, Fan; Tang, Hong; Shang, Shang; Mathiak, Klaus; Cong, Fengyu

Lehti tai sarja: Applied Sciences

eISSN: 2076-3417

Julkaisuvuosi: 2020

Volyymi: 10

Lehden numero: 11

Artikkelinumero: 3956

Kustantaja: MDPI

Julkaisumaa: Sveitsi

Julkaisun kieli: englanti

DOI: https://doi.org/10.3390/app10113956

Avoin saatavuus: Open access -julkaisukanavassa ilmestynyt julkaisu

Julkaisu on rinnakkaistallennettu (JYX): https://jyx.jyu.fi/handle/123456789/69902


Tiivistelmä

Heart sounds play an important role in the diagnosis of cardiac conditions. Due to the low signal-to-noise ratio (SNR), it is problematic and time-consuming for experts to discriminate different kinds of heart sounds. Thus, objective classification of heart sounds is essential. In this study, we combined a conventional feature engineering method with deep learning algorithms to automatically classify normal and abnormal heart sounds. First, 497 features were extracted from eight domains. Then, we fed these features into the designed convolutional neural network (CNN), in which the fully connected layers that are usually used before the classification layer were replaced with a global average pooling layer to obtain global information about the feature maps and avoid overfitting. Considering the class imbalance, the class weights were set in the loss function during the training process to improve the classification algorithm’s performance. Stratified five-fold cross-validation was used to evaluate the performance of the proposed method. The mean accuracy, sensitivity, specificity and Matthews correlation coefficient observed on the PhysioNet/CinC Challenge 2016 dataset were 86.8%, 87%, 86.6% and 72.1% respectively. The proposed algorithm’s performance achieves an appropriate trade-off between sensitivity and specificity.


YSO-asiasanat: sydäntaudit; diagnostiikka; koneoppiminen; tiedonlouhinta; neuroverkot

Vapaat asiasanat: automatic heart sound classification; feature engineering; convolutional neural network


Liittyvät organisaatiot


OKM-raportointi: Kyllä

Raportointivuosi: 2020


Viimeisin päivitys 2020-18-08 klo 13:00