Translating amyloid PET of different radiotracers by a deep generative model for interchangeability

It is challenging to compare amyloid PET images obtained with different radiotracers. Here, we introduce a new approach to improve the interchangeability of amyloid PET acquired with different radiotracers through image-level translation. Deep generative networks were developed using unpaired PET datasets, consisting of 203 [11C]PIB and 850 [18F]florbetapir brain PET images. Using 15 paired PET datasets, the standardized uptake value ratio (SUVR) values obtained from pseudo-PIB or pseudo-florbetapir PET images translated using the generative networks was compared to those obtained from the original images. The generated amyloid PET images showed similar distribution patterns with original amyloid PET of different radiotracers. The SUVR obtained from the original [18F]florbetapir PET was lower than those obtained from the original [11C]PIB PET. The translated amyloid PET images reduced the difference in SUVR. The SUVR obtained from the pseudo-PIB PET images generated from [18F]florbetapir PET showed a good agreement with those of the original PIB PET (ICC = 0.87 for global SUVR). The SUVR obtained from the pseudo-florbetapir PET also showed a good agreement with those of the original [18F]florbetapir PET (ICC = 0.85 for global SUVR). The ICC values between the original and generated PET images were higher than those between original [11C]PIB and [18F]florbetapir images (ICC = 0.65 for global SUVR). Our approach provides the image-level translation of amyloid PET images obtained using different radiotracers. It may facilitate the clinical studies designed with variable amyloid PET images due to long-term clinical follow-up as well as multicenter trials by enabling the translation of different types of amyloid PET.

[1]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Dean F. Wong,et al.  Standardization of amyloid quantitation with florbetapir standardized uptake value ratios to the Centiloid scale , 2018, Alzheimer's & Dementia.

[3]  Hiroharu Kawanaka,et al.  Automatic disease stage classification of glioblastoma multiforme histopathological images using deep convolutional neural network , 2018, Biomedical engineering letters.

[4]  Stefan Klöppel,et al.  Asymmetries of amyloid-β burden and neuronal dysfunction are positively correlated in Alzheimer's disease. , 2015, Brain : a journal of neurology.

[5]  Robert A. Koeppe,et al.  The Centiloid Project: Standardizing quantitative amyloid plaque estimation by PET , 2015, Alzheimer's & Dementia.

[6]  Philip Bachman,et al.  Augmented CycleGAN: Learning Many-to-Many Mappings from Unpaired Data , 2018, ICML.

[7]  Dong Soo Lee,et al.  Generation of Structural MR Images from Amyloid PET: Application to MR-Less Quantification , 2017, The Journal of Nuclear Medicine.

[8]  Si Eun Kim,et al.  Amyloid involvement in subcortical regions predicts cognitive decline , 2018, European Journal of Nuclear Medicine and Molecular Imaging.

[9]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[10]  Jae Sung Lee.  A Review of Deep-Learning-Based Approaches for Attenuation Correction in Positron Emission Tomography , 2021, IEEE Transactions on Radiation and Plasma Medical Sciences.

[11]  Keith A. Johnson,et al.  PET staging of amyloidosis using striatum , 2018, Alzheimer's & Dementia.

[12]  Ranjan Duara,et al.  Phase 3 trial of flutemetamol labeled with radioactive fluorine 18 imaging and neuritic plaque density. , 2015, JAMA neurology.

[13]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[14]  Dong Young Lee,et al.  Adaptive template generation for amyloid PET using a deep learning approach , 2018, Human brain mapping.

[15]  K. McGraw,et al.  Forming inferences about some intraclass correlation coefficients. , 1996 .

[16]  Karl J. Friston,et al.  Unified segmentation , 2005, NeuroImage.

[17]  Yuichi Yoshida,et al.  Spectral Normalization for Generative Adversarial Networks , 2018, ICLR.

[18]  Jae Sung Lee,et al.  Computed tomography super-resolution using deep convolutional neural network , 2018, Physics in medicine and biology.

[19]  M. Mintun,et al.  Amyloid-β Imaging with Pittsburgh Compound B and Florbetapir: Comparing Radiotracers and Quantification Methods , 2013, The Journal of Nuclear Medicine.

[20]  Paul Babyn,et al.  Generative Adversarial Network in Medical Imaging: A Review , 2018, Medical Image Anal..

[21]  M. Pontecorvo,et al.  Amyloid imaging in Alzheimer's disease: comparison of florbetapir and Pittsburgh compound-B positron emission tomography , 2012, Journal of Neurology, Neurosurgery & Psychiatry.

[22]  Sina Honari,et al.  Distribution Matching Losses Can Hallucinate Features in Medical Image Translation , 2018, MICCAI.

[23]  John Ashburner,et al.  A fast diffeomorphic image registration algorithm , 2007, NeuroImage.

[24]  Jae Sung Lee,et al.  Improving the Accuracy of Simultaneously Reconstructed Activity and Attenuation Maps Using Deep Learning , 2018, The Journal of Nuclear Medicine.

[25]  W. Jagust,et al.  The Alzheimer's Disease Neuroimaging Initiative 2 PET Core: 2015 , 2015, Alzheimer's & Dementia.

[26]  C. Geula,et al.  Asymmetry and heterogeneity of Alzheimer's and frontotemporal pathology in primary progressive aphasia. , 2014, Brain : a journal of neurology.

[27]  Yaozong Gao,et al.  Estimating CT Image from MRI Data Using 3D Fully Convolutional Networks , 2016, LABELS/DLMIA@MICCAI.

[28]  C. Rowe,et al.  Comparison of 11C-PiB and 18F-florbetaben for Aβ imaging in ageing and Alzheimer’s disease , 2012, European Journal of Nuclear Medicine and Molecular Imaging.

[29]  N. Tzourio-Mazoyer,et al.  Automated Anatomical Labeling of Activations in SPM Using a Macroscopic Anatomical Parcellation of the MNI MRI Single-Subject Brain , 2002, NeuroImage.

[30]  Karl Herholz,et al.  Clinical amyloid imaging in Alzheimer's disease , 2011, The Lancet Neurology.

[31]  John Seibyl,et al.  Florbetaben PET imaging to detect amyloid beta plaques in Alzheimer's disease: Phase 3 study , 2015, Alzheimer's & Dementia.

[32]  Christian Igel,et al.  Deep Feature Learning for Knee Cartilage Segmentation Using a Triplanar Convolutional Neural Network , 2013, MICCAI.

[33]  R. Coleman,et al.  Use of Florbetapir-PET for Imaging-Amyloid Pathology , 2011 .

[34]  R. Coleman,et al.  Use of florbetapir-PET for imaging beta-amyloid pathology. , 2011, JAMA.

[35]  拓海 杉山,et al.  “Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告 , 2017 .

[36]  Jung-Woo Ha,et al.  StarGAN: Unified Generative Adversarial Networks for Multi-domain Image-to-Image Translation , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[37]  W. Klunk,et al.  Imaging brain amyloid in Alzheimer's disease with Pittsburgh Compound‐B , 2004, Annals of neurology.

[38]  Min Hyoung Cho,et al.  U-net based metal segmentation on projection domain for metal artifact reduction in dental CT , 2019, Biomedical engineering letters.