Adversarial cycle-consistent synthesis of cerebral microbleeds for data augmentation

We propose a novel framework for controllable pathological image synthesis for data augmentation. Inspired by CycleGAN, we perform cycle-consistent imageto-image translation between two domains: healthy and pathological. Guided by a semantic mask, an adversarially trained generator synthesizes pathology on a healthy image in the specified location. We demonstrate our approach on an institutional dataset of cerebral microbleeds in traumatic brain injury patients. We utilize synthetic images generated with our method for data augmentation in cerebral microbleeds detection. Enriching the training dataset with synthetic images exhibits the potential to increase detection performance for cerebral microbleeds in traumatic brain injury patients.

[1]  Steven Warach,et al.  Cerebral Microbleeds : A Field Guide to their Detection and Interpretation , 2012 .

[2]  P Kapeller,et al.  MRI evidence of past cerebral microbleeds in a healthy elderly population , 1999, Neurology.

[3]  Hayit Greenspan,et al.  An Adversarial Learning Approach to Medical Image Synthesis for Lesion Detection , 2018, IEEE Journal of Biomedical and Health Informatics.

[4]  Alexander Zelinsky,et al.  Fast Radial Symmetry for Detecting Points of Interest , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  B. M. ter Haar Romeny,et al.  Automated detection of cerebral microbleeds in patients with Traumatic Brain Injury , 2016, NeuroImage: Clinical.

[6]  Saeed Hassanpour,et al.  Generative Image Translation for Data Augmentation in Colorectal Histopathology Images , 2019, ML4H@NeurIPS.

[7]  Jeffrey L. Gunter,et al.  Medical Image Synthesis for Data Augmentation and Anonymization using Generative Adversarial Networks , 2018, SASHIMI@MICCAI.

[8]  Anant Gupta,et al.  Generative Image Translation for Data Augmentation of Bone Lesion Pathology , 2018, MIDL.

[9]  Brian B. Avants,et al.  The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) , 2015, IEEE Transactions on Medical Imaging.

[10]  Saifeng Liu,et al.  Cerebral microbleed detection using Susceptibility Weighted Imaging and deep learning , 2019, NeuroImage.

[11]  Mohsen Ghafoorian,et al.  Cerebral microbleed detection in traumatic brain injury patients using 3D convolutional neural networks , 2018, Medical Imaging.

[12]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[13]  David J. Werring,et al.  Cerebral Microbleeds: Pathophysiology to Clinical Practice , 2011 .

[14]  Joaquim Salvi,et al.  Multiple Sclerosis Lesion Synthesis in MRI Using an Encoder-Decoder U-NET , 2019, IEEE Access.

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

[16]  Aaron C. Courville,et al.  Improved Training of Wasserstein GANs , 2017, NIPS.

[17]  S. Tsaftaris,et al.  Pseudo-healthy synthesis with pathology disentanglement and adversarial learning , 2020, Medical Image Anal..