论文信息 - Brain Tumor Classification Using Convolutional Neural Network

Brain Tumor Classification Using Convolutional Neural Network

Medical image classification has gained tremendous attention in recent years, and Convolutional Neural Network (CNN) is the most widespread neural network model for image classification problem. CNN is designed to determine features adaptively through backpropagation by applying numerous building blocks, such as convolution layers, pooling layers, and fully connected layers. In this paper, we mainly focused on developing a CNN model for classifying brain tumors in T1-weighted contrast-enhanced MRI images. The proposed system consists of two significant steps. First, preprocess the images using different image processing techniques and then classify the preprocessed image using CNN. The experiment is conducted on a dataset of 3064 images which contain three types of brain tumor (glioma, meningioma, pituitary). We achieved a high testing accuracy of 94.39%, average precision of 93.33% and an average recall of 93% using our CNN model. The proposed system exhibited satisfying accuracy on the dataset and outperformed many of the prominent existing methods.

Sunanda Das | O. F. M. Riaz Rahman Aranya | Nishat Nayla Labiba

[1] Yang Gao,et al. Content-Based Image Retrieval Using Spatial Layout Information in Brain Tumor T1-Weighted Contrast-Enhanced MR Images , 2014, PloS one.

[2] M. Zweig,et al. Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. , 1993, Clinical chemistry.

[3] Jun Zhang,et al. Implementation of Training Convolutional Neural Networks , 2015, ArXiv.

[4] J. Barnholtz-Sloan,et al. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2009-2013. , 2016, Neuro-oncology.

[5] Jun Cheng,et al. brain tumor dataset , 2016 .

[6] Jimmy Ba,et al. Adam: A Method for Stochastic Optimization , 2014, ICLR.

[7] Nitish Srivastava,et al. Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[8] J. Barnholtz-Sloan,et al. CBTRUS statistical report: primary brain and central nervous system tumors diagnosed in the United States in 2007-2011. , 2012, Neuro-oncology.

[9] Yang Gao,et al. Content-based retrieval of brain tumor in contrast-enhanced MRI images using tumor margin information and learned distance metric. , 2012, Medical physics.

[10] Qianjin Feng,et al. Retrieval of Brain Tumors with Region-Specific Bag-of-Visual-Words Representations in Contrast-Enhanced MRI Images , 2012, Comput. Math. Methods Medicine.