Machine learning in quantitative PET: A review of attenuation correction and low-count image reconstruction methods

• Type of Material: Article
• Authors: Tonghe Wang, Emory University; Yang Lei, Emory University; Yabo Fu, Emory University; Walter J Curran, Emory University; Tian Liu, Emory University; Jonathon Nye, Emory University; Xiaofeng Yang, Emory University
• Language: English
• Date: 2020-08-01
• Publisher: ELSEVIER SCI LTD
• Publication Version: Author Accepted Manuscript (After Peer Review)
• Copyright Statement: © 2020 Associazione Italiana di Fisica Medica. Published by Elsevier Ltd. All rights reserved.
• License: Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International
• Final Published Version (URL): http://dx.doi.org/10.1016/j.ejmp.2020.07.028
• Title of Journal or Parent Work: PHYSICA MEDICA-EUROPEAN JOURNAL OF MEDICAL PHYSICS
• Volume: 76
• Start Page: 294
• End Page: 306
• Grant/Funding Information: This research was supported in part by the National Cancer Institute of the National Institutes of Health under Award Number R01CA215718 and Emory Winship Cancer Institute pilot grant.
• Keywords: FDG PET/CT; LYMPH-NODE METASTASES; Machine learning; CT IMAGES; POSITRON-EMISSION-TOMOGRAPHY; CONVOLUTIONAL NEURAL-NETWORKS; Positron emission tomography; PARTIAL-VOLUME CORRECTION; PET; Radiology, Nuclear Medicine & Medical Imaging; Low-count PET; WHOLE-BODY PET; F-18-FDG PET; GROSS TUMOR VOLUME; Life Sciences & Biomedicine; Science & Technology; Attenuation correction; LUNG-CANCER

Abstract

The rapid expansion of machine learning is offering a new wave of opportunities for nuclear medicine. This paper reviews applications of machine learning for the study of attenuation correction (AC) and low-count image reconstruction in quantitative positron emission tomography (PET). Specifically, we present the developments of machine learning methodology, ranging from random forest and dictionary learning to the latest convolutional neural network-based architectures. For application in PET attenuation correction, two general strategies are reviewed: 1) generating synthetic CT from MR or non-AC PET for the purposes of PET AC, and 2) direct conversion from non-AC PET to AC PET. For low-count PET reconstruction, recent deep learning-based studies and the potential advantages over conventional machine learning-based methods are presented and discussed. In each application, the proposed methods, study designs and performance of published studies are listed and compared with a brief discussion. Finally, the overall contributions and remaining challenges are summarized.