Real-Time Detection of Meningiomas by Image Segmentation: A Very Deep Transfer Learning Convolutional Neural Network Approach

Developing a treatment strategy that effectively prolongs the lives of people with brain tumours requires an accurate diagnosis of the condition. Machine learning has made great strides thanks to the development of convolutional neural networks and computer-aided tumour detection systems. The deep convolutional layers automatically extract important and dependable information from the input space, in contrast to more traditional neural network layers. One recent and promising advancement in this field is machine learning. Still, there is a dearth of study being done in this area. Therefore, starting with the analysis of magnetic resonance images, we have suggested in this research work a tried-and-true and methodical strategy for real-time meningioma diagnosis by image segmentation using a Very Deep Transfer Learning CNN Model or DNN Model (VGG-16) with CUDA. Since it produces a greater level of accuracy than other deep CNN models like AlexNet, GoogleNet, etc., we have chosen to employ the VGGNet CNN (Convolutional Neural Network) model. The VGG network that we have constructed with very small convolutional filters consists of 13 convolutional layers and 3 fully connected layers. Our VGGNet model takes in an sMRI FLAIR image input. VGG’s convolutional layers leverage a minimal receptive field, i.e., 3×3, the smallest possible size that still captures up/down and left/right. Moreover, there are also 1×1 convolution filters acting as a linear transformation of the input. This is followed by a ReLU unit. The convolution stride is fixed at 1 pixel to keep the spatial resolution preserved after convolution. All the hidden layers in our VGG network also use ReLU. A dataset consisting of 264 3D FLAIR sMRI image segments from all three different classes (meningioma, tuberculoma, and normal) was employed. The number of epochs in the Sequential Model has been set to 10. The Keras layers that we have used are Dense, Dropout, Flatten, Batch Normalization, ReLU. According to the simulation findings, our suggested model successfully classified all of the data in our used dataset with a 99.0% overall accuracy. The performance metrics of the implemented model and confusion matrix for tumor classification indicate the model’s high accuracy in brain tumor classification. The good outcomes demonstrate the possibility of our suggested method as a useful diagnostic, better understanding and prognostic tool for clinical outcomes, and efficient brain tumour treatment planning tool. It was unraveled that the several performance metrics we computed using the confusion matrix of the previously used model are very good. Consequently, we think that the approach we have suggested is a great way to identify brain tumours.