Self-trained deep convolutional neural network for noise reduction in CT

Zhongxing Zhou, Akitoshi Inoue, Cynthia H. McCollough, Lifeng Yu

Research output: Contribution to journalArticlepeer-review

Abstract

Purpose: Supervised deep convolutional neural network (CNN)-based methods have been actively used in clinical CT to reduce image noise. The networks of these methods are typically trained using paired high- and low-quality data from a massive number of patients and/or phantom images. This training process is tedious, and the network trained under a given condition may not be generalizable to patient images acquired and reconstructed under different conditions. We propose a self-trained deep CNN (ST-CNN) method for noise reduction in CT that does not rely on preexisting training datasets. Approach: The ST-CNN training was accomplished using extensive data augmentation in the projection domain, and the inference was applied to the data itself. Specifically, multiple independent noise insertions were applied to the original patient projection data to generate multiple realizations of low-quality projection data. Then, rotation augmentation was adopted for both the original and low-quality projection data by applying the rotation angle directly on the projection data so that images were rotated at arbitrary angles without introducing additional bias. A large number of paired low- and high-quality images from the same patient were reconstructed and paired for training the ST-CNN model. Results: No significant difference was found between the ST-CNN and conventional CNN models in terms of the peak signal-to-noise ratio and structural similarity index measure. The ST-CNN model outperformed the conventional CNN model in terms of noise texture and homogeneity in liver parenchyma as well as better subjective visualization of liver lesions. The ST-CNN may sacrifice the sharpness of vessels slightly compared to the conventional CNN model but without affecting the visibility of peripheral vessels or diagnosis of vascular pathology. Conclusions: The proposed ST-CNN method trained from the data itself may achieve similar image quality in comparison with conventional deep CNN denoising methods pre-trained on external datasets.

Original languageEnglish (US)
Article number044008
JournalJournal of Medical Imaging
Volume10
Issue number4
DOIs
StatePublished - Jul 1 2023

Keywords

  • CT
  • denoise
  • self-trained
  • supervised deep convolutional neural network

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Fingerprint

Dive into the research topics of 'Self-trained deep convolutional neural network for noise reduction in CT'. Together they form a unique fingerprint.

Cite this