의학물리학

본문글자크기
  • [J Nucl Med.] Generation of PET Attenuation Map for Whole-Body Time-of-Flight 18F-FDG PET/MRI Using a Deep Neural Network Trained with Simultaneously Reconstructed Activity and Attenuation Maps.

    서울의대 / 황동휘, 이재성*

  • 출처
    J Nucl Med.
  • 등재일
    2019 Aug
  • 저널이슈번호
    60(8):1183-1189. doi: 10.2967/jnumed.118.219493. Epub 2019 Jan 25.
  • 내용

    바로가기  >

    Abstract
    We propose a new deep learning-based approach to provide more accurate whole-body PET/MRI attenuation correction than is possible with the Dixon-based 4-segment method. We use activity and attenuation maps estimated using the maximum-likelihood reconstruction of activity and attenuation (MLAA) algorithm as inputs to a convolutional neural network (CNN) to learn a CT-derived attenuation map. Methods: The whole-body 18F-FDG PET/CT scan data of 100 cancer patients (38 men and 62 women; age, 57.3 ± 14.1 y) were retrospectively used for training and testing the CNN. A modified U-net was trained to predict a CT-derived μ-map (μ-CT) from the MLAA-generated activity distribution (λ-MLAA) and μ-map (μ-MLAA). We used 1.3 million patches derived from 60 patients' data for training the CNN, data of 20 others were used as a validation set to prevent overfitting, and the data of the other 20 were used as a test set for the CNN performance analysis. The attenuation maps generated using the proposed method (μ-CNN), μ-MLAA, and 4-segment method (μ-segment) were compared with the μ-CT, a ground truth. We also compared the voxelwise correlation between the activity images reconstructed using ordered-subset expectation maximization with the μ-maps, and the SUVs of primary and metastatic bone lesions obtained by drawing regions of interest on the activity images. Results: The CNN generates less noisy attenuation maps and achieves better bone identification than MLAA. The average Dice similarity coefficient for bone regions between μ-CNN and μ-CT was 0.77, which was significantly higher than that between μ-MLAA and μ-CT (0.36). Also, the CNN result showed the best pixel-by-pixel correlation with the CT-based results and remarkably reduced differences in activity maps in comparison to CT-based attenuation correction. Conclusion: The proposed deep neural network produced a more reliable attenuation map for 511-keV photons than the 4-segment method currently used in whole-body PET/MRI studies.

     


    Author information

    Hwang D1,2, Kang SK1,2, Kim KY1,2, Seo S3, Paeng JC2,4, Lee DS5,4,6, Lee JS7,2,4.
    1
    Department of Biomedical Sciences, Seoul National University, Seoul, Korea.
    2
    Department of Nuclear Medicine, Seoul National University, Seoul, Korea.
    3
    Department of Neuroscience, College of Medicine, Gachon University, Incheon, Korea jaes@snu.ac.kr.
    4
    Institute of Radiation Medicine, Medical Research Center, Seoul National University, Seoul, Korea; and.
    5
    Department of Nuclear Medicine, Seoul National University, Seoul, Korea jaes@snu.ac.kr.
    6
    Department of Molecular Medicine and Biopharmaceutical Sciences, Graduate School of Convergence Science and Technology, Seoul National University, Suwon, Korea.
    7
    Department of Biomedical Sciences, Seoul National University, Seoul, Korea jaes@snu.ac.kr.

  • 키워드
    PET/MRI; attenuation correction; deep learning; simultaneous reconstruction
  • 편집위원

    deep neural network를 이용해 기존보다 더 향상된 attenuation map을 PET/MRI에 적용한 점이 인상깊음

    2019-09-27 15:09:43

  • 덧글달기
    덧글달기
       IP : 54.145.183.34

    등록