Self-supervised Noise2noise Method Utilizing Corrupted Images with a Modular Network for LDCT Denoising
This addresses the problem of limited paired data for medical imaging denoising, offering an incremental improvement by enabling training with only noisy images.
The paper tackles LDCT image denoising without requiring paired noisy-clean datasets by proposing a self-supervised method that uses only LDCT data, combining noise2noise and noisy-as-clean strategies with a modular U-Net, and shows effectiveness on the Mayo LDCT dataset compared to state-of-the-art methods.
Deep learning is a very promising technique for low-dose computed tomography (LDCT) image denoising. However, traditional deep learning methods require paired noisy and clean datasets, which are often difficult to obtain. This paper proposes a new method for performing LDCT image denoising with only LDCT data, which means that normal-dose CT (NDCT) is not needed. We adopt a combination including the self-supervised noise2noise model and the noisy-as-clean strategy. First, we add a second yet similar type of noise to LDCT images multiple times. Note that we use LDCT images based on the noisy-as-clean strategy for corruption instead of NDCT images. Then, the noise2noise model is executed with only the secondary corrupted images for training. We select a modular U-Net structure from several candidates with shared parameters to perform the task, which increases the receptive field without increasing the parameter size. The experimental results obtained on the Mayo LDCT dataset show the effectiveness of the proposed method compared with that of state-of-the-art deep learning methods. The developed code is available at https://github.com/XYuan01/Self-supervised-Noise2Noise-for-LDCT.