Synthetic 18F-FDG PET Image Generation Using a Combination of Biomathematical Modeling and Machine Learning
by
, , and
Mohammad Amin Abazari
1
,
Madjid Soltani
1,2,3,4,5,*,
Farshad Moradi Kashkooli
1 and
Kaamran Raahemifar
2,6,7
1
Department of Mechanical Engineering, K. N. Toosi University of Technology, Tehran 19967-15433, Iran
2
Faculty of Science, School of Optometry and Vision Science, University of Waterloo, Waterloo, ON N2L 3G1, Canada
3
Advanced Bioengineering Initiative Center, Multidisciplinary International Complex, K. N. Toosi Univesity of Technology, Tehran 14176-14411, Iran
4
Center for Biotechnology and Bioengineering (CBB), University of Waterloo, Waterloo, ON N2L 3G1, Canada
5
Department of Electrical and Computer Engineering, Faculty of Engineering, University of Waterloo, Waterloo, ON N2L 3G1, Canada
6
Data Science and Artificial Intelligence Program, College of Information Sciences and Technology (IST), Penn State University, State College, PA 16801, USA
7
Department of Chemical Engineering, University of Waterloo, 200 University Avenue West, Waterloo, ON N2L 3G1, Canada
*
Author to whom correspondence should be addressed.
Cancers 2022, 14(11), 2786; https://doi.org/10.3390/cancers14112786
Submission received: 25 April 2022
/
Revised: 21 May 2022
/
Accepted: 1 June 2022
/
Published: 3 June 2022
(This article belongs to the Collection Artificial Intelligence in Oncology)
Simple Summary
Training computer-assisted algorithms on medical images, particularly 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET) due to its excellent diagnostic accuracy, is difficult, considering small/fragmented samples or privacy concerns. In computer-vision, deep learning-based models, unlike the conventional data augmentation methods, are highly sought after for creating massive medical samples. For this reason, we developed a multi-scale computational framework to generate synthetic 18F-FDG PET images similar to the real ones in different stages of solid tumor growth and angiogenesis. The framework is developed based on the bio-physiological phenomena of FDG radiotracer uptake in solid tumors using a biomathematical model and a generative adversarial network (GAN)-based architecture. The non-invasive augmented 18F-FDG PET images can be used in clinical practice without the need to manage the patient data. In addition, our spatiotemporal mathematical model can calculate the distribution of various radiopharmaceuticals in different tumor-associated vasculatures.
Abstract
No previous works have attempted to combine generative adversarial network (GAN) architectures and the biomathematical modeling of positron emission tomography (PET) radiotracer uptake in tumors to generate extra training samples. Here, we developed a novel computational model to produce synthetic 18F-fluorodeoxyglucose (18F-FDG) PET images of solid tumors in different stages of progression and angiogenesis. First, a comprehensive biomathematical model is employed for creating tumor-induced angiogenesis, intravascular and extravascular fluid flow, as well as modeling of the transport phenomena and reaction processes of 18F-FDG in a tumor microenvironment. Then, a deep convolutional GAN (DCGAN) model is employed for producing synthetic PET images using 170 input images of 18F-FDG uptake in each of 10 different tumor microvascular networks. The interstitial fluid parameters and spatiotemporal distribution of 18F-FDG uptake in tumor and healthy tissues have been compared against previously published numerical and experimental studies, indicating the accuracy of the model. The structural similarity index measure (SSIM) and peak signal-to-noise ratio (PSNR) of the generated PET sample and the experimental one are 0.72 and 28.53, respectively. Our results demonstrate that a combination of biomathematical modeling and GAN-based augmentation models provides a robust framework for the non-invasive and accurate generation of synthetic PET images of solid tumors in different stages.
Keywords:
synthetic medical image generation; medical image augmentation; deep convolutional generative adversarial networks (DCGANs); [18F]-fluorodeoxyglucose positron emission tomography (18F-FDG PET) imaging; multi-scale computational cancer modeling; tumor-induced angiogenesis; kinetic modeling of diagnostic agents; spatiotemporal modeling
