Share Email Print
cover

Proceedings Paper

Population and individual information guided PET image denoising using deep neural network
Author(s): Jianan Cui; Kuang Gong; Ning Guo; Chenxi Wu; Kyungsang Kim; Huafeng Liu; Quanzheng Li
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Positron emission tomography (PET) images still suffer from low signal-to-noise ratio (SNR) due to various physical degradation factors. Recently deep neural networks (DNNs) have been successfully applied to medical image denoising tasks when large number of training pairs are available. Previously the deep image prior framework1 shows that individual information can be enough to train a denoising network, with noisy image itself as the training label. In this work, we propose to improve PET image quality by jointly employing population and individual information based on DNN. The population information was utilized by pre-training the network using a group of patients. The individual information was introduced during testing phase by fine-tuning the population-information-trained network. Unlike traditional DNN denoising, in this framework fine-tuning during testing phase is available as the noisy PET image itself was treated as the training label. Quantification results based on clinical PET/MR datasets containing thirty patients demonstrate that the proposed framework outperforms Gaussian, non-local mean and deep image prior denoising methods.

Paper Details

Date Published: 28 May 2019
PDF: 5 pages
Proc. SPIE 11072, 15th International Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine, 110721E (28 May 2019); doi: 10.1117/12.2534901
Show Author Affiliations
Jianan Cui, Zhejiang Univ. (China)
Massachusetts General Hospital (United States)
Kuang Gong, Massachusetts General Hospital (United States)
Harvard Medical School (United States)
Ning Guo, Massachusetts General Hospital (United States)
Harvard Medical School (United States)
Chenxi Wu, Massachusetts General Hospital (United States)
Kyungsang Kim, Massachusetts General Hospital (United States)
Harvard Medical School (United States)
Huafeng Liu, Zhejiang Univ. (China)
Quanzheng Li, Massachusetts General Hospital (United States)
Harvard Medical School (United States)


Published in SPIE Proceedings Vol. 11072:
15th International Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine
Samuel Matej; Scott D. Metzler, Editor(s)

© SPIE. Terms of Use
Back to Top
PREMIUM CONTENT
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?
close_icon_gray