Automated Diagnosis of Brain Tumor Based on Deep Learning Feature Fusion Using MRI Images
Kondepudi Venkata Durga, Debendra Muduli, Kumar Rahul, Amballa Vijay Sai Charan Naidu, Majji Jayanth Kumar, Santosh Kumar Sharma
Abstract
Brain tumor detection is an important task in medical image analysis, as early detection is crucial for the patient's treatment and survival. In recent years, deep learning has shown remarkable success in various medical imaging tasks, including brain tumor detection. The proposed model is based on the deep learning feature fusion of two pre-trained models called Inception V3and VGG19. In this work, we compare the performance of 8 pre-trained Convolutional Neural Network (CNN) models using ImageNet dataset weights in order to identify the best suitable model. The experimental work has been evaluated on a publicly available MRI dataset. The proposed model achieves the greatest accuracy of 96% as compared to other predefined deep learning models. We used the Adam optimizer and also evaluated the performance of this combined model using various evaluation metrics, including accuracy, precision, recall, and F1 score. This study demonstrates the potential of deep learning in medical image analysis and can help clinicians in the early detection of brain tumors.