Computer-Aided Diagnosis System for Automated Detection of Mri Brain Tumors
Umar S. Alqasemi1, Sultan A. Almutawa2, Shadi M. Obaid3
1Umar S. Alqasemi, Department Electrical and Computer Engineering, King Abdulaziz University, Jeddah, Makkah, Saudi Arabia.
2Sultan A. Almutawa, Department Electrical and Computer Engineering, King Abdulaziz University, Jeddah, Makkah, Saudi Arabia.
3Shadi M. Obaid, Department Electrical and Computer Engineering, King Abdulaziz University, Jeddah, Makkah, Saudi Arabia.
Manuscript received on 02 August 2023 | Revised Manuscript received on 19 January 2024 | Manuscript Accepted on 15 February 2024 | Manuscript published on 28 February 2024 | PP: 40-38 | Volume-13 Issue-3, February 2024 | Retrieval Number: 100.1/ijeat.C436013030224 | DOI: 10.35940/ijeat.C4360.13030224
Open Access | Editorial and Publishing Policies | Cite | Zenodo | OJS | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Detection and classification of brain tumors in manual or traditional way is an area which could be improved by having such automated detection and clarification system for brain tumors. In this paper, enhanced Computer-Aided Diagnosis CAD software system is introduced for brain tumor detection and classification. Total of 229 brain MRI images was taken as dataset for the purpose of this research; those dataset images include 105 normal brain MRI images, and 124 abnormal brain MRI images. Proposed CAD system is specialized for Meningioma brain tumor detection and classification, and the technique could be generalized and implemented for Glioma, and Pituitary brain tumors as well, and the whole system was implemented using MATLAB software. We started by cropping the region of interest (ROI) of dataset images. Then, feature extraction was implemented using first order statistical features, as well as using of some wavelets filters in combination with the former. T-test is used to exclude features of no statistical significance (p-value < 0.05). After that, different types of classifiers were used to separate the normal set from the abnormal one. Note that, we used an iterative approach to by changing features with many runs until we got best performance, where, best accuracy results were gotten with SVM-Kernel Function (Linear), KNN-1, KNN-3, and KNN-5 classifiers. Note also that, we used convolutional neural networks (CNN) from Deep Learning toolbox of MATLAB as a control method to compare, where the images were fed directly to the CNN. The results were evaluated using performance assessment techniques which are Sensitivity, Specificity, Positive Predictive Value (PPV), Negative Predictive Value (NPV), Accuracy, Error Rate, and Area Under the Curve (AUC) of Reciever Operator Characteristic (ROC). With SVM classifier, the best gotten accuracy results were 91 % with CNN classifier, 82% with SVM classifier, and 77 % with KNN classifier. Furthermore, it was very beneficial to find such feature extraction techniques which gave acceptable accuracy results with three different classifiers; this was the case two times as mentioned the study. All proposed CAD system areas was developed and implemented using MATLAB software.
Keywords: CAD System, MATLAB, Automated Detection and Classification, MRI Brain Tumors, SVM Classifier, KNN Classifier, CNN, DCNN
Scope of the Article: Classification