A Novel U-Net Architecture for Denoising of Real-world Noise Corrupted Phonocardiogram Signal

The article presents a novel U-Net based deep neural network architecture for denoising phonocardiogram (PCG) signals contaminated by real-world noise. The proposed architecture aims to improve the diagnosis process by removing the noises that corrupt the heart sound signals, such as lung sound, coughing, and sneezing. The performance of the proposed denoising technique has shown improvement compared to the existing state-of-the-art denoising algorithms.

Publication date: 4 Oct 2023
Project Page: Not provided
Paper: https://arxiv.org/pdf/2310.00216