Technical note: Phantom-based training framework for convolutional neural network CT noise reduction.

Journal: Medical physics
PMID:

Abstract

BACKGROUND: Deep artificial neural networks such as convolutional neural networks (CNNs) have been shown to be effective models for reducing noise in CT images while preserving anatomic details. A practical bottleneck for developing CNN-based denoising models is the procurement of training data consisting of paired examples of high-noise and low-noise CT images. Obtaining these paired data are not practical in a clinical setting where the raw projection data is not available. This work outlines a technique to optimize CNN denoising models using methods that are available in a routine clinical setting.

Authors

  • Nathan R Huber
    From the Department of Radiology, Mayo Clinic, Rochester, MN.
  • Andrew D Missert
    CT Clinical Innovation Center, Department of Radiology, Mayo Clinic, Rochester, MN, 55905, USA.
  • Hao Gong
    Department of Radiology, Mayo Clinic, Rochester, MN, 55905, USA.
  • Shuai Leng
    Department of Radiology, Mayo Clinic, Rochester, MN, 55905, USA.
  • Lifeng Yu
    Hithink RoyalFlush Information Network Co., Ltd., Hangzhou 310023, China. yulifeng@myhexin.com.
  • Cynthia H McCollough
    Department of Radiology, Mayo Clinic, Rochester, MN, 55905, USA.