nach oben

Erschienen in:

20.04.2020 | Short Communication

AI approach of cycle-consistent generative adversarial networks to synthesize PET images to train computer-aided diagnosis algorithm for dementia

verfasst von: Yuichi Kimura, Aya Watanabe, Takahiro Yamada, Shogo Watanabe, Takashi Nagaoka, Mitsutaka Nemoto, Koichi Miyazaki, Kohei Hanaoka, Hayato Kaida, Kazunari Ishii

Erschienen in: Annals of Nuclear Medicine | Ausgabe 7/2020

Einloggen, um Zugang zu erhalten

Abstract

Objective

An artificial intelligence (AI)-based algorithm typically requires a considerable amount of training data; however, few training images are available for dementia with Lewy bodies and frontotemporal lobar degeneration. Therefore, this study aims to present the potential of cycle-consistent generative adversarial networks (CycleGAN) to obtain enough number of training images for AI-based computer-aided diagnosis (CAD) algorithms for diagnosing dementia.

Methods

We trained CycleGAN using 43 amyloid-negative and 45 positive images in slice-by-slice.

Results

The CycleGAN can be used to synthesize reasonable amyloid-positive images, and the continuity of slices was preserved.

Discussion

Our results show that CycleGAN has the potential to generate a sufficient number of training images for CAD of dementia.

Lu D, Popuri K, Ding GW, Balachandar R, Beg MF. Alzheimer’s disease neuroimaging initiative. Multiscale deep neural network based analysis of FDG-PET images for the early diagnosis of Alzheimer’s disease. Med Image Anal. 2018;46:26–34.CrossRef

Van Ginneken B, Schaefer-Prokop CM, Prokop M. Computer-aided diagnosis: how to move from the laboratory to the clinic. Radiology. 2011;261(3):719–32.CrossRef

Radford A, Metz L, Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks. Int Conf Learning Represent 2016 (ICLR2016) 2016; 1–16. Xiv: 1511.06434v2.

Wang Y, Yu B, Wang L, Zu C, Lalush DS, Lin W, Wu X, Zhou J, Shen D, Zhou L. 3D conditional generative adversarial networks for high-quality PET image estimation at low dose. NeuroImage. 2018;174:550–62.CrossRef

Zhu JY, Park T, Isola P, Efros AA. Unpaired image-to-image translation using cycle-consistent adversarial networks. Proc IEEE Int Conf Comput Vis 2017;2223–2232. arXiv: 1703.10593v6.

Hiasa Y, Otake Y, Takao M, Matsuoka T, Takashima K, Carass A, Prince JL, Sugano N, Sato Y. Cross-modality image synthesis from unpaired data using CycleGAN. In: Int Workshop Simul Synthesis Med Imag. Lecture Notes in Computer Science; vol. 11037. Cham: Springer; 2018. p. 31–41.

Yi X, Walia E, Babyn P. Generative adversarial network in medical imaging: a review. Med Image Anal. 2019. https://doi.org/10.1016/j.media.2019.101552.CrossRefPubMed

Hosokawa C, Ishii K, Kimura Y, Hyodo T, Hosono M, Sakaguchi K, Usami K, Shimamoto K, Yamazoe Y, Murakami T. Performance of ¹¹C-Pittsburgh Compound B PET binding potential images in the detection of amyloid deposits on equivocal static images. J Nucl Med. 2015;56(12):1910–5.CrossRef

Scheiwein FT, Ishii K, Hosokawa C, Kaida H, Hyodo T, Hanaoka K, Brendel M, Bartenstein P, Rominger A, Murakami T. Regional differences in amyloid deposition between 11C-Pib PET positive patients with and without elevated striatal amyloid uptake. J Alzheimers Dis Parkinsonism. 2017;7(2):1–7. https://doi.org/10.4172/2161-0460.1000317.CrossRef

10.

Chinzei K, Shimizu A, Mori K, Harada K, Takeda H, Hashidume M, Ishizuka M, Kato N, Kawamori R, Kyo S, Nagata K, Yamane T, Sakuma I, Ohe K, Mitsuishi M. Regulatory science on AI-based medical devices and systems. Adv Biomed Eng. 2018;7:118–23.CrossRef

Titel: AI approach of cycle-consistent generative adversarial networks to synthesize PET images to train computer-aided diagnosis algorithm for dementia
verfasst von: Yuichi Kimura
Aya Watanabe
Takahiro Yamada
Shogo Watanabe
Takashi Nagaoka
Mitsutaka Nemoto
Koichi Miyazaki
Kohei Hanaoka
Hayato Kaida
Kazunari Ishii
Publikationsdatum: 20.04.2020
Verlag: Springer Singapore
Erschienen in: Annals of Nuclear Medicine / Ausgabe 7/2020
Print ISSN: 0914-7187
Elektronische ISSN: 1864-6433
DOI: https://doi.org/10.1007/s12149-020-01468-5

Live-Webinar: Aktuelle Leitlinien bei Herz-Kreislauf-Erkrankungen

Springer Medizin

AI approach of cycle-consistent generative adversarial networks to synthesize PET images to train computer-aided diagnosis algorithm for dementia

Abstract

Objective

Methods

Results

Discussion

Live-Webinar: Aktuelle Leitlinien bei Herz-Kreislauf-Erkrankungen

Springer Medizin

Abstract

Objective

Methods

Results

Discussion

Bitte loggen Sie sich ein, um Zugang zu diesem Inhalt zu erhalten

Weitere Artikel der Ausgabe 7/2020

An impact of microscopic positive margin on incomplete response after I-131 treatment in differentiated thyroid cancer

Oral 67Gallium Citrate Scintigraphy in assessment of inflammatory activity of Crohn’s disease: methodological issues on accuracy and agreement

Synthesis and evaluation of novel radioiodinated anthranilate derivatives for in vivo imaging of vascular endothelial growth factor receptor with single-photon emission computed tomography

In vitro/in vivo assessment of the targeting ability of [99mTc] Tc-labeled an aptide specific to the extra domain B of fibronectin (APTEDB) for colorectal cancer

Methods for analyzing binary of data clusters into intestinal segments of the patients

Improvement in cognitive decline after indirect bypass surgery in adult moyamoya disease: implication of 15O-gas positron emission tomography