Hippocampal subfields segmentation in brain MR images using generative adversarial networks

Abstract

BackgroundSegmenting the hippocampal subfields accurately from brain magnetic resonance (MR) images is a challenging task in medical image analysis. Due to the small structural size and the morphological complexity of the hippocampal subfields, the traditional segmentation methods are hard to obtain the ideal segmentation result.MethodsIn this paper, we proposed a hippocampal subfields segmentation method using generative adversarial networks. The proposed method can achieve the pixel-level classification of brain MR images by building an UG-net model and an adversarial model and training the two models against each other alternately. UG-net extracts local information and retains the interrelationship features between pixels. Moreover, the adversarial training implements spatial consistency among the generated class labels and smoothens the edges of class labels on segmented region.ResultsThe evaluation has performed on the dataset obtained from center for imaging of neurodegenerative diseases (CIND) for CA1, CA2, DG, CA3, Head, Tail, SUB, ERC and PHG in hippocampal subfields, resulting in the dice similarity coefficient (DSC) of 0.919, 0.648, 0.903, 0.673, 0.929, 0.913, 0.906, 0.884 and 0.889 respectively. For the large subfields, such as Head and CA1 of hippocampus, the DSC was increased by 3.9% and 9.03% than state-of-the-art approaches, while for the smaller subfields, such as ERC and PHG, the segmentation accuracy was significantly increased 20.93% and 16.30% respectively.ConclusionThe results show the improvement in performance of the proposed method, compared with other methods, which include approaches based on multi-atlas, hierarchical multi-atlas, dictionary learning and sparse representation and CNN. In implementation, the proposed method provides better results in hippocampal subfields segmentation.

Enhancing magnetic resonance imaging-driven Alzheimer’s disease classification performance using generative adversarial learning

“…Since its introduction, there has been a surge of interest in the application of GAN frameworks related to the brain. Some of the applications include image generation with improved properties such as achieving super resolution or better quality [6][7][8][9][10][11], data augmentation [12][13][14], segmentation [9,[13][14][15][16], image reconstruction [17][18][19][20], image-to-image translation [21][22][23][24], and motion correction [25,26]. While these important studies have demonstrated the exciting prospect of using GAN architectures, there is a limited amount of work that has focused on utilizing the generated images for subsequent tasks such as disease classification [27].…”

A novel deep learning based hippocampus subfield segmentation method

“…Recently, deep learning based methods has been also proposed for hippocampus subfield segmentation. For example, UGNET has been proposed (Shi et al, 2019) using an adversarial training approach and also variants of the famous UNET architecture (Ronneberger et al, 2015) such as the Dilated Dense UNET (Zhu et al,2019) have been proposed. However, one of the major problems of supervised deep learning methods is their hunger for training data to be able to generalize on unseen data.…”

Combining residual attention mechanisms and generative adversarial networks for hippocampus segmentation

“…The combination of the generative and adversarial models enhances the spatial continuity of the segmentation results and improves the accuracy of segmentation. Shi et al [30] proposed a method based on GAN to achieve the hippocampus segmentation. This method uses the UNet network as a generative model and performs interactive training with the adversarial model to achieve pixel-level classification of brain MRI images.…”