FDCNet: Presentation of the Fuzzy CNN and Fractal Feature Extraction for Detection and Classification of Tumors
Open Access
- 6 May 2022
- journal article
- research article
- Published by Wiley in Computational Intelligence and Neuroscience
- Vol. 2022, 1-16
Abstract
The detection of brain tumors using magnetic resonance imaging is currently one of the biggest challenges in artificial intelligence and medical engineering. It is important to identify these brain tumors as early as possible, as they can grow to death. Brain tumors can be classified as benign or malignant. Creating an intelligent medical diagnosis system for the diagnosis of brain tumors from MRI imaging is an integral part of medical engineering as it helps doctors detect brain tumors early and oversee treatment throughout recovery. In this study, a comprehensive approach to diagnosing benign and malignant brain tumors is proposed. The proposed method consists of four parts: image enhancement to reduce noise and unify image size, contrast, and brightness, image segmentation based on morphological operators, feature extraction operations including size reduction and selection of features based on the fractal model, and eventually, feature improvement according to segmentation and selection of optimal class with a fuzzy deep convolutional neural network. The BraTS data set is used as magnetic resonance imaging data in experimental results. A series of evaluation criteria is also compared with previous methods, where the accuracy of the proposed method is 98.68%, which has significant results.Keywords
This publication has 53 references indexed in Scilit:
- Role of magnetic resonance spectroscopy & diffusion weighted imaging in differentiation of supratentorial brain tumorsEgyptian Journal of Radiology and Nuclear Medicine, 2016
- Tumor infiltrating immune cells in gliomas and meningiomasBrain, Behavior, and Immunity, 2016
- Tumor Detection In Brain Using Genetic AlgorithmProcedia Computer Science, 2016
- An unsupervised learning method with a clustering approach for tumor identification and tissue segmentation in magnetic resonance brain imagesApplied Soft Computing, 2016
- Performance analysis of classifier for brain tumor detection and diagnosisComputers and Electrical Engineering, 2015
- Brain tumors detection and segmentation in MR images: Gabor wavelet vs. statistical featuresComputers and Electrical Engineering, 2015
- Improved Edge Detection Algorithm for Brain Tumor SegmentationProcedia Computer Science, 2015
- Fuzzy anisotropic diffusion based segmentation and texture based ensemble classification of brain tumorApplied Soft Computing, 2014
- Improving MR brain image segmentation using self-organising maps and entropy-gradient clusteringInformation Sciences, 2014
- Intelligent Detection of Intrusion into Databases Using Extended Classifier System (XCS)International Journal of Electrical and Computer Engineering (IJECE), 2013