Neural network-based quantitative reconstruction of PET without attenuation correction

Linlin Zhao; Huafeng Liu

doi:10.1117/12.2602183

9 October 2021 Neural network-based quantitative reconstruction of PET without attenuation correction

Linlin Zhao, Huafeng Liu

Proceedings Volume 11900, Optics in Health Care and Biomedical Optics XI; 1190012 (2021) https://doi.org/10.1117/12.2602183
Event: SPIE/COS Photonics Asia, 2021, Nantong, JS, China

Abstract

The importance of accurate attenuation correction (AC) in positron emission tomography (PET) has been widely recognized. However AC based PET/CT or PET/MR suffers from several problems such as artifacts and noises. In this paper, we propose a novel method based on physics-driven Iteration Shrinkage-Thresholding Algorithm (PISTA) to achieve quantitative PET image reconstruction from raw sinogram data without AC. The PISTA utilizes system matrix which is used as geometry of projection between quantitative PET activity images and sinograms without AC, to generate the next updated quantitative PET activity images as the input of learnable part. An evaluation study on simulated phantom data is described, where experimental results have shown great promise for such strategy.

Conference Presentation

Citation Download Citation

Linlin Zhao and Huafeng Liu "Neural network-based quantitative reconstruction of PET without attenuation correction", Proc. SPIE 11900, Optics in Health Care and Biomedical Optics XI, 1190012 (9 October 2021); https://doi.org/10.1117/12.2602183

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