0 406

Cited 0 times in

PET/CT for Brain Amyloid: A Feasibility Study for Scan Time Reduction by Deep Learning

 Sangwon Lee  ;  Jin Ho Jung  ;  Dongwoo Kim  ;  Hyun Keong Lim  ;  Mi-Ae Park  ;  Garam Kim  ;  Minjae So  ;  Sun Kook Yoo  ;  Byoung Seok Ye  ;  Yong Choi  ;  Mijin Yun 
 CLINICAL NUCLEAR MEDICINE, Vol.46(3) : e133-e140, 2021-03 
Journal Title
Issue Date
Amyloid / metabolism ; Aniline Compounds ; Brain / diagnostic imaging ; Brain / metabolism ; Deep Learning* ; Feasibility Studies ; Female ; Humans ; Image Processing, Computer-Assisted / methods* ; Positron Emission Tomography Computed Tomography* ; Retrospective Studies ; Signal-To-Noise Ratio ; Stilbenes
Purpose: This study was to develop a convolutional neural network (CNN) model with a residual learning framework to predict the full-time 18F-florbetaben (18F-FBB) PET/CT images from corresponding short-time scans.

Methods: In this retrospective study, we enrolled 22 cognitively normal subjects, 20 patients with mild cognitive impairment, and 42 patients with Alzheimer disease. Twenty minutes of list-mode PET/CT data were acquired and reconstructed as the ground-truth images. The short-time scans were made in either 1, 2, 3, 4, or 5 minutes. The CNN with a residual learning framework was implemented to predict the ground-truth images of 18F-FBB PET/CT using short-time scans with either a single-slice or a 3-slice input layer. Model performance was evaluated by quantitative and qualitative analyses. Additionally, we quantified the amyloid load in the ground-truth and predicted images using the SUV ratio.

Results: On quantitative analyses, with increasing scan time, the normalized root-mean-squared error and the SUV ratio differences between predicted and ground-truth images gradually decreased, and the peak signal-to-noise ratio increased. On qualitative analysis, the predicted images from the 3-slice CNN model showed better image quality than those from the single-slice model. The 3-slice CNN model with a short-time scan of at least 2 minutes achieved comparable, quantitative prediction of full-time 18F-FBB PET/CT images, with adequate to excellent image quality.

Conclusions: The 3-slice CNN model with a residual learning framework is promising for the prediction of full-time 18F-FBB PET/CT images from short-time scans.
Full Text
Appears in Collections:
1. College of Medicine (의과대학) > Dept. of Medical Engineering (의학공학교실) > 1. Journal Papers
1. College of Medicine (의과대학) > Dept. of Neurology (신경과학교실) > 1. Journal Papers
1. College of Medicine (의과대학) > Dept. of Nuclear Medicine (핵의학교실) > 1. Journal Papers
Yonsei Authors
Kim, Dongwoo(김동우) ORCID logo https://orcid.org/0000-0002-1723-604X
Ye, Byoung Seok(예병석) ORCID logo https://orcid.org/0000-0003-0187-8440
Yoo, Sun Kook(유선국) ORCID logo https://orcid.org/0000-0002-6032-4686
Yun, Mi Jin(윤미진) ORCID logo https://orcid.org/0000-0002-1712-163X
사서에게 알리기


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.