ABSTRACT
Preoperative glioma grading is important to patient's follow-up treatment and prognosis. In recent years, deep learning technology has developed rapidly in computer vision. Various medical images have enabled deep learning to be widely used in computer-aided diagnosis (CAD). Typically, these methods fuse multi-modal image information or multi-model structures. They only use the encoder's final feature map representation output for classification. In this paper, ResNet and graph convolutional network (GCN) are cascaded. Then the hierarchical feature map of the encoder is used to construct the graph topology. Finally, GCN analyzes the graph topology to achieve case-level glioma classification. Our model achieved an F1-score of 0.907 and an average test accuracy of 93.74%. It outperforms the traditional convolutional neural network (CNN) model.
- Geert Litjens, Thijs Kooi, Babak Ehteshami Bejnordi, Arnaud Arindra Adiyoso Setio, Francesco Ciompi, Mohsen Ghafoorian, Jeroen A.W.M. van der Laak, Bram van Ginneken, and Clara I. Sánchez. 2017. A survey on deep learning in medical image analysis. Medical Image Analysis 42 (December 2017), 60–88. https://doi.org/10.1016/j.media.2017.07.005Google ScholarCross Ref
- Kirti Raj Bhatele and Sarita Singh Bhadauria. 2021. Machine learning application in Glioma classification: review and comparison analysis. Archives of Computational Methods in Engineering 29 (April 2021), 247-274. https://doi.org/10.1007/s11831-021-09572-zGoogle ScholarCross Ref
- Xiaohong W. Gao, Rui Hui, and Zengmin Tian. 2017. Classification of CT brain images based on deep learning networks. Computer Methods and Programs in Biomedicine 138 (January 2017), 49-56. https://doi.org/10.1016/j.cmpb.2016.10.007Google ScholarDigital Library
- Salma Alqazzaz, Xianfang Sun, Xin Yang, and Len Nokes. 2019. Automated brain tumor segmentation on multi-modal MR image using SegNet. Computational Visual Media 5 (April 2019), 209-219. https://doi.org/10.1007/s41095-019-0139-yGoogle ScholarCross Ref
- Mengxin Liu and Jun Yang. 2021. Image Classification of Brain tumor based on Channel Attention Mechanism. Journal of Physics: Conference Series 2035, 1 (September, 2021), 20-29. https://dx.doi.org/10.1088/1742-6596/2035/1/012029Google ScholarCross Ref
- Linmin Pei, Lasitha Vidyaratne, Md Monibor Rahman, and Khan M. Iftekharuddin. 2020. Context aware deep learning for brain tumor segmentation, subtype classification, and survival prediction using radiology images. Scientific Reports 10, 1 (November 2020), 1-11. https://doi.org/10.1038/s41598-020-74419-9Google ScholarCross Ref
- Michael London and Michael Häusser. 2005. Dendritic Computation. Annual Review of Neuroscience 28 (July 2005), 503-532. https://doi.org/10.1146/annurev.neuro.28.061604.135703Google ScholarCross Ref
- Gang Liu and Jing Wang. 2022. Dendrite Net: A White-Box Module for Classification, Regression, and System Identification. IEEE Transactions on Cybernetics 52, 12 (December 2022), 13774-13787. https://ieeexplore.ieee.org/document/9619767Google ScholarCross Ref
- Bjoern H. Menze, Andras Jakab, Stefan Bauer, Jayashree Kalpathy-Cramer, Keyvan Farahani, and Justin Kirby, 2015. The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS). IEEE Transactions on Medical Imaging 34, 10 (October 2015), 1993-2024. https://ieeexplore.ieee.org/document/6975210Google ScholarCross Ref
- Imperial College London, South Kensington Campus. 2014. IXI Dataset – Brain Development. Retrieved May 15, 2023 from https://brain-development.org/ixi-dataset/Google Scholar
Index Terms
- Preoperative Glioma Grading Based on Hierarchical Information Using Residual Network and Graph Convolutional Network
Recommendations
Automatic Breast Cancer Grading of Histological Images using Dilated Residual Network
ICBBT '19: Proceedings of the 2019 11th International Conference on Bioinformatics and Biomedical TechnologyBreast cancer is one of the leading causes of female death worldwide. Histological evaluation of the breast biopsies is essential in the early detection. Recently, deep learning methods are developed to automatically grade breast cancer of histological ...
TriResNet: A Deep Triple-Stream Residual Network for Histopathology Grading
Image Analysis and RecognitionAbstractWhile microscopic analysis of histopathological slides is generally considered as the gold standard method for performing cancer diagnosis and grading, the current method for analysis is extremely time consuming and labour intensive as it requires ...
A computer-aided grading of glioma tumor using deep residual networks fusion
Highlights- A novel deep learning-based glioma grading framework has been proposed.
- The ...
AbstractBackground and objectives: Among different cancer types, glioma is considered as a potentially fatal brain cancer that arises from glial cells. Early diagnosis of glioma helps the physician in offering effective ...
Comments