Dual-tracer PET image direct reconstruction and separation based on three-dimensional encoder-decoder network

Fuzhen Zeng; Huafeng Liu

doi:10.1117/12.2573439

10 October 2020 Dual-tracer PET image direct reconstruction and separation based on three-dimensional encoder-decoder network

Fuzhen Zeng, Huafeng Liu

Author Affiliations +

Proceedings Volume 11553, Optics in Health Care and Biomedical Optics X; 115530X (2020) https://doi.org/10.1117/12.2573439
Event: SPIE/COS Photonics Asia, 2020, Online Only

Abstract

Dual-tracer PET signal separation is a typical problem in the field of PET imaging. Previous studies mainly focused on separating dual-tracer PET signals on activity image reconstructed by traditional iterative methods, which ignore the influence of quality of reconstructed images and need additional time for reconstruction pro- cedure. In this work, dual-tracer PET signal separation is cracked as simultaneously direct reconstruction and separation from mixed sinogram based on deep learning, we introduced a three-dimensional encoder-decoder net- work to achieve it. The network can learn and combine temporal information and spatial information properly, and spatiotemporal information plays an important role in signal separation and structure reconstruction respec- tively. We evaluated the proposed method both in Monte Carlo simulation experiment and SD rats experiment. Experimental results show that the proposed method can obtain better single tracer PET activity image than another method also based on deep learning and belongs to direct reconstruction.

Conference Presentation

Citation Download Citation

Fuzhen Zeng and Huafeng Liu "Dual-tracer PET image direct reconstruction and separation based on three-dimensional encoder-decoder network", Proc. SPIE 11553, Optics in Health Care and Biomedical Optics X, 115530X (10 October 2020); https://doi.org/10.1117/12.2573439

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available