Detection of atrial fibrillation via artificial intelligence-assisted auscultation

Figures

• 

  Fig. 1. Study flow chart

• 

  Fig. 2. Artificial intelligence digital stethoscope

• 

  Fig. 3. Data preprocessing, (a) Data Conversion, (b) Data Truncation, (c) Band-pass Filtering, (d) Downsampling, (e) Normalization, (f) Audio Segmentation

• 

  Fig. 4. Feature extraction pipeline and representative feature maps: a Short-Time Fourier Transform (STFT): Time-frequency representation with 256-point Hamming window, 50% overlap. b Mel-spectrogram (1024 points): Logarithmic frequency scale mimicking human auditory perception, computed with 1024-point Fourier transform. c Mel-frequency cepstral coefficients (MFCCs): 20 coefficients derived from the Mel-spectrogram using discrete cosine transform. d Bispectrum analysis: Bi-spectral analysis showing phase-coupled frequency components

• 

  Fig. 5. Heart sound monitoring and health management system

• 

  Fig. 6. Receiver operating characteristic curves of the model using the derivation and validation datasets