Volume 21 No 3 (2023)
 Download PDF
Automated Classification of Brain Tumor Images using Hybrid Machine Learning Techniques
N. Phani Bindu, P. Narahari Sastry
Brain tumor is one of the most common types of tumor, and it is one of the primary reasons of human death and also it has its effect on the humans of all age groups. But, with early detection, the severity of the tumor can be reduced, and there may be a chance of getting better treatment. Generally, brain tumor classification is done using biopsy, which can be achieved only after brain surgery and this method is invasive. So, different types of automated hybrid classification techniques can be utilized to reduce human errors during diagnosis by radiologists and avoid surgery. With advanced technology, i.e. by the hybrid methods obtained by combining machine learning, and deep learning algorithms, radiologists can easily diagnose the tumor from MRI Scans without any invasive methods. This paper introduces a hybrid method for classifying tumors and obtaining different grades of tumors. In order to use two classification algorithms with MRI images, this paper also utilizes a hybrid CNN architecture method. In this technique, feature extraction is done using AlexNet and VGG, while image classification is done using SVM. The main strategy combines the CNN algorithm's pre-trained AlexNet and VGG algorithms. The feature vector obtained is given to SVM and it is employed for pattern classification. The second method integrates a soft-max classifier with such a finely trained AlexNet.
A brain tumor; MRI images; deep learning; CNN; AlexNet; VGG; SVM.
Copyright © Neuroquantology

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Articles published in the Neuroquantology are available under Creative Commons Attribution Non-Commercial No Derivatives Licence (CC BY-NC-ND 4.0). Authors retain copyright in their work and grant IJECSE right of first publication under CC BY-NC-ND 4.0. Users have the right to read, download, copy, distribute, print, search, or link to the full texts of articles in this journal, and to use them for any other lawful purpose.