CASPIANET++: A multidimensional Channel-Spatial Asymmetric attention network with Noisy Student Curriculum Learning paradigm for brain tumor segmentation. (September 2021)

• Record Type: Journal Article
• Authors: Liew, Andrea; Lee, Chun Cheng; Lan, Boon Leong; Tan, Maxine
• Is Part Of: Computers in biology and medicine. Volume 136(2021)
• Publication Date: 2021-09
• Subjects: Convolutional neural networks -- Attention -- Brain tumor segmentation -- Self-training -- Curriculum learning; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2021.104690
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Convolutional neural networks (CNNs) have been used quite successfully for semantic segmentation of brain tumors. However, current CNNs and attention mechanisms are stochastic in nature and neglect the morphological indicators used by radiologists to manually annotate regions of interest. In this paper, we introduce a channel and spatial wise asymmetric attention (CASPIAN) by leveraging the inherent structure of tumors to detect regions of saliency. To demonstrate the efficacy of our proposed layer, we integrate this into a well-established convolutional neural network (CNN) architecture to achieve higher Dice scores, with less GPU resources. Also, we investigate the inclusion of auxiliary multiscale and multiplanar attention branches to increase the spatial context crucial in semantic segmentation tasks. The resulting architecture is the new CASPIANET++, which achieves Dice Scores of 91.19%, 87.6% and 81.03% for whole tumor, tumor core and enhancing tumor respectively. Furthermore, driven by the scarcity of brain tumor data, we investigate the Noisy Student method for segmentation tasks. Our new Noisy Student Curriculum Learning paradigm, which infuses noise incrementally to increase the complexity of the training images exposed to the network, further boosts the enhancing tumor region to 81.53%. Additional validation performed on the BraTS2020 data shows that the Noisy Student Curriculum Learning method works well without any additional training or finetuning. Highlights: Attention layer with multiscale and multiplanar channel-spatial scheme Channel-Spatial attention combines a-priori knowledge with stochastic methods Multiplanar attention evaluates spatial information from three planar axes Multiscale method compensates for the loss in spatial information in the encoder New curriculum learning paradigm improves segmentation accuracy in low data regimes