Literature DB >> 35169887

Direct and indirect strategies of deep-learning-based attenuation correction for general purpose and dedicated cardiac SPECT.

Xiongchao Chen1, Bo Zhou1, Huidong Xie1, Luyao Shi1, Hui Liu2,3, Wolfgang Holler4, MingDe Lin2,5, Yi-Hwa Liu6,7, Edward J Miller2,6, Albert J Sinusas1,2,6, Chi Liu8,9.   

Abstract

PURPOSE: Deep-learning-based attenuation correction (AC) for SPECT includes both indirect and direct approaches. Indirect approaches generate attenuation maps (μ-maps) from emission images, while direct approaches predict AC images directly from non-attenuation-corrected (NAC) images without μ-maps. For dedicated cardiac SPECT scanners with CZT detectors, indirect approaches are challenging due to the limited field-of-view (FOV). In this work, we aim to 1) first develop novel indirect approaches to improve the AC performance for dedicated SPECT; and 2) compare the AC performance between direct and indirect approaches for both general purpose and dedicated SPECT.
METHODS: For dedicated SPECT, we developed strategies to predict truncated μ-maps from NAC images reconstructed with a small matrix, or full μ-maps from NAC images reconstructed with a large matrix using 270 anonymized clinical studies scanned on a GE Discovery NM/CT 570c SPECT/CT. For general purpose SPECT, we implemented direct and indirect approaches using 400 anonymized clinical studies scanned on a GE NM/CT 850c SPECT/CT. NAC images in both photopeak and scatter windows were input to predict μ-maps or AC images.
RESULTS: For dedicated SPECT, the averaged normalized mean square error (NMSE) using our proposed strategies with full μ-maps was 1.20 ± 0.72% as compared to 2.21 ± 1.17% using the previous direct approaches. The polar map absolute percent error (APE) using our approaches was 3.24 ± 2.79% (R2 = 0.9499) as compared to 4.77 ± 3.96% (R2 = 0.9213) using direct approaches. For general purpose SPECT, the averaged NMSE of the predicted AC images using the direct approaches was 2.57 ± 1.06% as compared to 1.37 ± 1.16% using the indirect approaches.
CONCLUSIONS: We developed strategies of generating μ-maps for dedicated cardiac SPECT with small FOV. For both general purpose and dedicated SPECT, indirect approaches showed superior performance of AC than direct approaches.
© 2022. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

Entities:  

Keywords:  Attenuation correction; Dedicated SPECT; Deep learning; Myocardial perfusion imaging

Mesh:

Year:  2022        PMID: 35169887      PMCID: PMC9253078          DOI: 10.1007/s00259-022-05718-8

Source DB:  PubMed          Journal:  Eur J Nucl Med Mol Imaging        ISSN: 1619-7070            Impact factor:   10.057


  33 in total

1.  Quantification of nuclear cardiac images: the Yale approach.

Authors:  Yi-Hwa Liu
Journal:  J Nucl Cardiol       Date:  2007-07       Impact factor: 5.952

Review 2.  SPECT/CT physical principles and attenuation correction.

Authors:  James A Patton; Timothy G Turkington
Journal:  J Nucl Med Technol       Date:  2008-02-20

3.  Attenuation correction using deep learning for brain perfusion SPECT images.

Authors:  Kenta Sakaguchi; Hayato Kaida; Shuhei Yoshida; Kazunari Ishii
Journal:  Ann Nucl Med       Date:  2021-03-09       Impact factor: 2.668

4.  [Effect of Misregistration between SPECT and CT Images on Attenuation Correction for Quantitative Bone SPECT Imaging].

Authors:  Mika Sakoshi; Norikazu Matsutomo; Tomoaki Yamamoto; Eisuke Sato
Journal:  Nihon Hoshasen Gijutsu Gakkai Zasshi       Date:  2018

5.  The impact of system matrix dimension on small FOV SPECT reconstruction with truncated projections.

Authors:  Chung Chan; Joyoni Dey; Yariv Grobshtein; Jing Wu; Yi-Hwa Liu; Rachel Lampert; Albert J Sinusas; Chi Liu
Journal:  Med Phys       Date:  2016-01       Impact factor: 4.071

6.  Myocardial perfusion quantitation with 15O-labelled water PET: high reproducibility of the new cardiac analysis software (Carimas).

Authors:  Sergey V Nesterov; Chunlei Han; Maija Mäki; Sami Kajander; Alexandru G Naum; Hans Helenius; Irina Lisinen; Heikki Ukkonen; Mikko Pietilä; Esa Joutsiniemi; Juhani Knuuti
Journal:  Eur J Nucl Med Mol Imaging       Date:  2009-04-30       Impact factor: 9.236

7.  Prognostic study of risk stratification among Japanese patients with ischemic heart disease using gated myocardial perfusion SPECT: J-ACCESS study.

Authors:  Tsunehiko Nishimura; Kenichi Nakajima; Hideo Kusuoka; Akira Yamashina; Shigeyuki Nishimura
Journal:  Eur J Nucl Med Mol Imaging       Date:  2007-10-10       Impact factor: 9.236

8.  Artificial intelligence-based attenuation correction; closer to clinical reality?

Authors:  Robert J H Miller; Piotr J Slomka
Journal:  J Nucl Cardiol       Date:  2021-07-06       Impact factor: 3.872

9.  Independent attenuation correction of whole body [18F]FDG-PET using a deep learning approach with Generative Adversarial Networks.

Authors:  Karim Armanious; Tobias Hepp; Thomas Küstner; Helmut Dittmann; Konstantin Nikolaou; Christian La Fougère; Bin Yang; Sergios Gatidis
Journal:  EJNMMI Res       Date:  2020-05-24       Impact factor: 3.138

10.  A deep learning approach for 18F-FDG PET attenuation correction.

Authors:  Fang Liu; Hyungseok Jang; Richard Kijowski; Gengyan Zhao; Tyler Bradshaw; Alan B McMillan
Journal:  EJNMMI Phys       Date:  2018-11-12
View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.