Generative Model of Brain Microbleeds for MRI Detection of Vascular Marker of Neurodegenerative Diseases

Frontiers in Neuroscience

Leo Lebrat²  Paul Yates³  Olivier Salvado⁵  Amir Fazlollahi⁶  Yongsheng Gao⁷  Saba Momeni⁷  Alan Wee-Chung Liew⁸  Christopher Rowe⁹ 

[1] Commonwealth Scientific and Industrial Research Organisation (CSIRO) Data61, Brisbane, QLD, Australia;Commonwealth Scientific and Industrial Research Organisation (CSIRO) Health and Biosecurity, Australian E-Health Research Centre, Brisbane, QLD, Australia;Department of Geriatric Medicine, Austin Health, Heidelberg, VIC, Australia;Department of Molecular Imaging and Therapy, Austin Health, Heidelberg, VIC, Australia;Department of Nuclear Medicine, Centre for PET, Austin Health, Heidelberg, VIC, Australia;Queensland Brain Institute, The University of Queensland, Brisbane, QLD, Australia;School of Engineering and Built Environment, Griffith University, Nathan, QLD, Australia;School of Information and Communication Technology, Griffith University, Nathan, QLD, Australia;The Florey Department of Neuroscience and Mental Health, The University of Melbourne, Parkville, VIC, Australia;

关键词: generative adversarial network;    cerebral microbleed;    data augmentation;    deep learning;    SWI images;    synthetic data;   

DOI  :  10.3389/fnins.2021.778767

来源: DOAJ

【 摘 要 】

Cerebral microbleeds (CMB) are increasingly present with aging and can reveal vascular pathologies associated with neurodegeneration. Deep learning-based classifiers can detect and quantify CMB from MRI, such as susceptibility imaging, but are challenging to train because of the limited availability of ground truth and many confounding imaging features, such as vessels or infarcts. In this study, we present a novel generative adversarial network (GAN) that has been trained to generate three-dimensional lesions, conditioned by volume and location. This allows one to investigate CMB characteristics and create large training datasets for deep learning-based detectors. We demonstrate the benefit of this approach by achieving state-of-the-art CMB detection of real CMB using a convolutional neural network classifier trained on synthetic CMB. Moreover, we showed that our proposed 3D lesion GAN model can be applied on unseen dataset, with different MRI parameters and diseases, to generate synthetic lesions with high diversity and without needing laboriously marked ground truth.

【 授权许可】

Unknown