Utilizing deep learning techniques to improve image quality and noise reduction in preclinical low-dose PET images in the sinogram domain

被引:2
|
作者
Doss, Kishore Krishnagiri Manoj [1 ]
Chen, Jyh-Cheng [1 ,2 ,3 ,4 ]
机构
[1] Natl Yang Ming Chiao Tung Univ, Dept Biomed Imaging & Radiol Sci, Taipei, Taiwan
[2] China Med Univ, Dept Med Imaging & Radiol Sci, Taichung, Taiwan
[3] Xuzhou Med Univ, Sch Med Imaging, Xuzhou, Peoples R China
[4] Natl Yang Ming Chiao Tung Univ, Dept Biomed Imaging & Radiol Sci, 155 Li Nong St,Sec 2, Taipei 112, Taiwan
关键词
deep learning; generative adversarial network; low-dose image; positron emission tomography; preclinical imaging; sinogram;
D O I
10.1002/mp.16830
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
BackgroundLow-dose positron emission tomography (LD-PET) imaging is commonly employed in preclinical research to minimize radiation exposure to animal subjects. However, LD-PET images often exhibit poor quality and high noise levels due to the low signal-to-noise ratio. Deep learning (DL) techniques such as generative adversarial networks (GANs) and convolutional neural network (CNN) have the capability to enhance the quality of images derived from noisy or low-quality PET data, which encodes critical information about radioactivity distribution in the body.PurposeOur objective was to optimize the image quality and reduce noise in preclinical PET images by utilizing the sinogram domain as input for DL models, resulting in improved image quality as compared to LD-PET images.MethodsA GAN and CNN model were utilized to predict high-dose (HD) preclinical PET sinograms from the corresponding LD preclinical PET sinograms. In order to generate the datasets, experiments were conducted on micro-phantoms, animal subjects (rats), and virtual simulations. The quality of DL-generated images was weighted by performing the following quantitative measures: structural similarity index measure (SSIM), root mean squared error (RMSE), peak signal-to-noise ratio (PSNR), signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR). Additionally, DL input and output were both subjected to a spatial resolution calculation of full width half maximum (FWHM) and full width tenth maximum (FWTM). DL outcomes were then compared with the conventional denoising algorithms such as non-local means (NLM), block-matching, and 3D filtering (BM3D).ResultsThe DL models effectively learned image features and produced high-quality images, as reflected in the quantitative metrics. Notably, the FWHM and FWTM values of DL PET images exhibited significantly improved accuracy compared to LD, NLM, and BM3D PET images, and just as precise as HD PET images. The MSE loss underscored the excellent performance of the models, indicating that the models performed well. To further improve the training, the generator loss (G loss) was increased to a value higher than the discriminator loss (D loss), thereby achieving convergence in the GAN model.ConclusionsThe sinograms generated by the GAN network closely resembled real HD preclinical PET sinograms and were more realistic than LD. There was a noticeable improvement in image quality and noise factor in the predicted HD images. Importantly, DL networks did not fully compromise the spatial resolution of the images.
引用
收藏
页码:209 / 223
页数:15
相关论文
共 50 条
  • [31] Validation of Deep-Learning Image Reconstruction for Low-Dose Chest Computed Tomography Scan: Emphasis on Image Quality and Noise
    Kim, Joo Hee
    Yoon, Hyun Jung
    Lee, Eunju
    Kim, Injoong
    Cha, Yoon Ki
    Bak, So Hyeon
    KOREAN JOURNAL OF RADIOLOGY, 2021, 22 (01) : 131 - 138
  • [32] Estimating standard-dose PET from low-dose PET with deep learning
    Lei, Yang
    Dong, Xue
    Wang, Tonghe
    Higgins, Kristin
    Liu, Tian
    Curran, Walter J.
    Mao, Hui
    Nye, Jonathan A.
    Yang, Xiaofeng
    MEDICAL IMAGING 2020: IMAGE PROCESSING, 2021, 11313
  • [33] FDG dose reduction using deep learning-based image reconstruction on PET images
    Akita, Ryuji
    Takauchi, Komei
    Ishibashi, Mana
    Ono, Shogo
    Yokomachi, Kazushi
    Nishimaru, Eiji
    Kiguchi, Masao
    Awai, Kazuo
    JOURNAL OF NUCLEAR MEDICINE, 2024, 65
  • [34] Deep learning techniques in PET/CT imaging: A comprehensive review from sinogram to image space
    Fallahpoor, Maryam
    Chakraborty, Subrata
    Pradhan, Biswajeet
    Faust, Oliver
    Barua, Prabal Datta
    Chegeni, Hossein
    Acharya, Rajendra
    COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2024, 243
  • [35] A Model to Improve the Quality of Low-dose CT Scan Images
    Chircop, Francesca
    Debono, Carl James
    Bezzina, Paul
    Zarb, Francis
    2022 IEEE 21ST MEDITERRANEAN ELECTROTECHNICAL CONFERENCE (IEEE MELECON 2022), 2022, : 131 - 136
  • [36] Erratum to: A method to assess image quality for Low-dose PET: analysis of SNR, CNR, bias and image noise
    Jianhua Yan
    Josh Schaefferkoetter
    Maurizio Conti
    David Townsend
    Cancer Imaging, 16
  • [37] Image quality evaluation of real low-dose breast PET
    Satoh, Yoko
    Imai, Masamichi
    Ikegawa, Chihiro
    Onishi, Hiroshi
    JAPANESE JOURNAL OF RADIOLOGY, 2022, 40 (11) : 1186 - 1193
  • [38] Image quality evaluation of real low-dose breast PET
    Yoko Satoh
    Masamichi Imai
    Chihiro Ikegawa
    Hiroshi Onishi
    Japanese Journal of Radiology, 2022, 40 : 1186 - 1193
  • [39] Noise-Generating-Mechanism-Driven Unsupervised Learning for Low-Dose CT Sinogram Recovery
    Zeng, Dong
    Wang, Lei
    Geng, Mingrui
    Li, Sui
    Deng, Yun
    Xie, Qi
    Li, Danyang
    Zhang, Hua
    Li, Yuanqing
    Xu, Zongben
    Meng, Deyu
    Ma, Jianhua
    IEEE TRANSACTIONS ON RADIATION AND PLASMA MEDICAL SCIENCES, 2022, 6 (04) : 404 - 414
  • [40] Can Deep Learning Raise the Quality of Low-Dose CT Images Above That of Normal-Dose CT Images?
    Bai, T.
    Nguyen, D.
    Wang, G.
    Jiang, S.
    MEDICAL PHYSICS, 2020, 47 (06) : E585 - E585