Home About Login Current Archives Announcements Editorial Board
Submit Now For Authors Call for Submissions Statistics Contact
Home > Archives > Volume 20, No 8 (2022) > Article

DOI: 10.14704/nq.2022.20.8.NQ44732

Brain Tumor Detection by Incorporating Hyperparameter Optimization in Convolutional Neural Network

S. Shargunam, Dr. G. Rajakumar

Abstract

In modern days, recognition of brain tumors has ended up being a breathtaking challenge in scientific endeavors. Because of its common image quality and the fact that it does not require ionizing radiation, MRI is frequently used. Diagnosis of a brain tumor is an extremely troublesome undertaking for specialists to distinguish at the beginning period. The objective is to recognize the brain tumor from MRI images utilizing Image processing strategies. The proposed work incorporates Extraction to assess tumor to be the noteworthy class that would be glioma, meningioma, and pituitary. The brain tumor earnestness has been assessed using Convolutional Neural Network figuring which gives us exact results by playing out the hyperparameter tuning components. Experimental findings demonstrate the superiority of our profound learning approach to the conventional condition techniques. Optimizing the hyperparameters in Convolutional Neural Networks (CNN) takes a lot of time for many researchers and professionals. Experts must configure a collection of hyperparameter options using tuning techniques to obtain superior performance hyperparameters. The best results of this configuration are thereafter modeled and implemented in CNN. Using the grid search tuning strategy the best hyperparameter for the dataset has been found by comparing three Optimizers and Batch size, learning rate, and momentum for Hyperparameter tuning. The system's performance and accuracy are enhanced by fine-tuning the parameters. When compared to other hyperparameters, the best optimizer, stochastic gradient descent (SGD), with a batch size of 64 and a learning rate of 0.001, achieved the maximum accuracy of 78.21%.

Keywords

Deep learning, glioma tumor, neural networks, tumor detection, convolutional neural network, hyperparameter.

Full Text

PDF

References

?>