MENU

Click here to

×

Are you sure ?

Yes, do it No, cancel
  1. Home
  2. Meeting Program
  3. All Sessions
  4. All Imaging SNAP Oral
  5. Synthetic CT Generation From Non-Attenuation Corrected PET Images for Whole-Body PET Imaging

Synthetic CT Generation From Non-Attenuation Corrected PET Images for Whole-Body PET Imaging

X Dong1*, Y Lei1 , T Wang1 , K Higgins1,2 , T Liu1,2 , W Curran1,2 , H Mao2,3 , J Nye3 , X Yang1,2 , (1) Department of Radiation Oncology, Emory University, Atlanta, GA (2)Winship Cancer Institute, Emory University, Atlanta, GA (3)Department of Radiology and Imaging Sciences, Emory University, Atlanta, GA

Presentations

(Sunday, 7/14/2019) 1:00 PM - 2:00 PM

Room: 303

Purpose: Attenuation correction (AC) is an essential component of PET reconstruction, but faces challenges including inter-scan motion, extra radiation exposure, image artifacts such as truncation and distortion, erroneous transformation of structural voxel-intensities to PET mu-map values. We propose a deep-learning-based method to derive synthetic CT (SCT) images from non-attenuation corrected PET (NAC PET) images for attenuation correction on whole-body PET imaging.

Methods: A 3D cycle-consistent generative adversarial networks (cycle-GAN) framework was employed to synthesize CT images from NAC PET. The method learns a transformation that minimizes the difference between SCT, generated from NAC PET, and true CT. It also learns an inverse transformation such that cycle NAC PET image generated from the SCT is close to the real NAC PET image. Both generators are implemented by a fully convolution attention network (FCAN), and followed by a discriminator which structured as a fully convolutional network. By using FCAN, the generator can retrieve the most relevant information representing the relationship between NAC PET and CT. We conducted a retrospective study on 30 sets of whole-body PET/CT with leave-one-out cross validation. Mean absolute error (MAE) and peak signal-to-noise ratio (PSNR) were used to quantify the performance of the proposed method.

Results: The whole body SCT images generated with the proposed method demonstrate great resemblance to the real CT images. The mean MAE of the SCT over real CT is 91.0±11.1 Hounsfield Unit. The PSNR on the synthetic CT images is 24.7±0.84 dB.

Conclusion: We proposed a deep-learning-based approach to generate synthetic CT from NAC PET. SCT generated with proposed method shows great similarity to true CT images both qualitatively and quantitatively, which could be used for PET attenuation correction in the absence of structural information for whole-body PET imaging.

Keywords

Not Applicable / None Entered.

Taxonomy

Not Applicable / None Entered.

Contact Email

Additional Links