AUTHOR=Xiao Yuteng , Yin Hongsheng , Wang Shui-Hua , Zhang Yu-Dong 

TITLE=TReC: Transferred ResNet and CBAM for Detecting Brain Diseases

JOURNAL=Frontiers in Neuroinformatics

VOLUME=Volume 15 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/neuroinformatics/articles/10.3389/fninf.2021.781551

DOI=10.3389/fninf.2021.781551

ISSN=1662-5196

ABSTRACT=Early diagnosis of pathological brains leads to early interventions in brain diseases, which may help control the illness conditions, prolong the life of patients and even cure them. Therefore, the classification of brain diseases is a challenging but helpful task. However, it is hard to collect brain images, and the superabundance of images is also a great challenge for computing resources. Therefore, The paper proposes a new approach named TReC, a specific model for small-scale samples, to detect brain diseases based on magnetic resonance imaging (MRI). At first, the Residual Networks (ResNet) model, pre-trained on the ImageNet dataset, serves as initialisation. Subsequently, a simple attention mechanism named Convolutional Block Attention Module (CBAM) is introduced and added into every ResNet residual block. At the same time, the Fully Connected (FC) layers of the ResNet are replaced with new FC layers, which meet the goal of classification. Finally, all the parameters of our model, including the ResNet, the CBAM, and new FC layers, are retrained. The effectiveness of the proposed model is evaluated on brain MR datasets for multi-class and two-class tasks. Compared with other state-of-the-art models, our model reaches the best performance for two-class and multi-class tasks on brain diseases.