Deep Convolutional Neural Network With a Multi-Scale Attention Feature Fusion Module for Segmentation of Multimodal Brain Tumor

Abstract

As a non-invasive, low-cost medical imaging technology, magnetic resonance imaging (MRI) has become an important tool for brain tumor diagnosis. Many scholars have carried out some related researches on MRI brain tumor segmentation based on deep convolutional neural networks, and have achieved good performance. However, due to the large spatial and structural variability of brain tumors and low image contrast, the segmentation of MRI brain tumors is challenging. Deep convolutional neural networks often lead to the loss of low-level details as the network structure deepens, and they cannot effectively utilize the multi-scale feature information. Therefore, a deep convolutional neural network with a multi-scale attention feature fusion module (MAFF-ResUNet) is proposed to address them. The MAFF-ResUNet consists of a U-Net with residual connections and a MAFF module. The combination of residual connections and skip connections fully retain low-level detailed information and improve the global feature extraction capability of the encoding block. Besides, the MAFF module selectively extracts useful information from the multi-scale hybrid feature map based on the attention mechanism to optimize the features of each layer and makes full use of the complementary feature information of different scales. The experimental results on the BraTs 2019 MRI dataset show that the MAFF-ResUNet can learn the edge structure of brain tumors better and achieve high accuracy.

Multi-task learning architecture for brain tumor detection and segmentation in MRI images

“…It uses the up-sampling layers to recover the original feature map. 25 The decoder part reconstructs the mask based on its learning using upsampling and the features from the skip connections coming from the encoder side. 27 The task-specific decoder has four convolutional layers and four up-sampling layers to effectively reconstruct the image with the most negligible error.…”

“…This strategy preserves the majority of the image content within the cropped area. It does not affect the datasets, but it reduces the image size and computational complexity, improving the network’s performance 25 . It also helps in dealing with the class-imbalance problem to some extent as we effectively tried to reduce the black background.…”

“…The decoder takes the encoding and is responsible for reconstructing it into a complete image. It uses the up-sampling layers to recover the original feature map 25 . The decoder part reconstructs the mask based on its learning using upsampling and the features from the skip connections coming from the encoder side 27 .…”

Unsupervised Black-Box Model Domain Adaptation for Brain Tumor Segmentation

“…Semantic segmentation provides the pixel-wise annotation of lesions or anatomical structures and has been an important prerequisite for early diagnosis and treatment planning (Liu et al, 2020a , b ; He et al, 2022 ). Because of the high cost of manual delineations, there is a large demand for automatic segmentation tools for clinical practice.…”

“…For the past several years, with the development of data-driven deep learning, the performance of segmentation tasks has been substantially improved (Liu et al, 2020c , 2021h ). For example, U-Net and its follow-up backbones achieved outstanding performance compared with their predecessors, in many natural and medical image analysis tasks, including the brain tumor localization and segmentation from magnetic resonance (MR) images (MRI) (Liu et al, 2020d ; He et al, 2022 ).…”