Deep Convolutional Neural Network with Transfer Learning for Automatic Brain Tumor Detection from MRI
- Resource Type
- Conference
- Authors
- Mathew, Josmy; Srinivasan, N
- Source
- 2022 International Conference on Computing, Communication, Security and Intelligent Systems (IC3SIS) Computing, Communication, Security and Intelligent Systems (IC3SIS), 2022 International Conference on. :1-6 Jun, 2022
- Subject
- Communication, Networking and Broadcast Technologies
Computing and Processing
Robotics and Control Systems
Deep learning
Training
Magnetic resonance imaging
Computational modeling
Transfer learning
Brain modeling
Security
Deep Learning
Transfer Learning
Brain Tumor
Classification
Artificial Intelligence
- Language
Early detection of brain tumors improves treatment options and increases the patient’s survival percentage. Tumor cells are exceedingly difficult to detect due to their diversity. Manual brain tumors detection for cancer diagnosis from a large number of MRI images obtained in clinical practice is a challenging and time-consuming task. The primary goal of this study is to investigate the ability of various pre-trained DCNN models using transfer learning to detect diseased brain images. This work focuses on constructing an effective and reliable method for detecting brain tumors using MRI to aid neurologists in making quick and appropriate decisions. The framework, which includes a dataset of 3064 images has been created and applied for the identification of brain tumors employing deep learning techniques. We have used conceptual model with Transfer Learning (TL) to verify it and got 98.28% accuracy, 97.51% recall and 97.43% precision. The proposed TL based brain tumor detection algorithm outperforms existing pre-trained models.