[Paper Review] Advancing Brain Tumor Detection: A Thorough Investigation of CNNs, Clustering, and SoftMax Classification in the Analysis of MRI Images

Jonayet Miah, Duc Minh Cao+4 more|arXiv (Cornell University)|Oct 26, 2023

Brain Tumor Detection and Classification8 citations

TL;DR

The paper investigates CNN-based brain tumor detection on MRI data, comparing SoftMax, RBF, and Decision Tree classifiers, and introduces clustering-based feature extraction to improve accuracy, with SoftMax achieving up to 99.52% on test data.

ABSTRACT

Brain tumors pose a significant global health challenge due to their high prevalence and mortality rates across all age groups. Detecting brain tumors at an early stage is crucial for effective treatment and patient outcomes. This study presents a comprehensive investigation into the use of Convolutional Neural Networks (CNNs) for brain tumor detection using Magnetic Resonance Imaging (MRI) images. The dataset, consisting of MRI scans from both healthy individuals and patients with brain tumors, was processed and fed into the CNN architecture. The SoftMax Fully Connected layer was employed to classify the images, achieving an accuracy of 98%. To evaluate the CNN's performance, two other classifiers, Radial Basis Function (RBF) and Decision Tree (DT), were utilized, yielding accuracy rates of 98.24% and 95.64%, respectively. The study also introduced a clustering method for feature extraction, improving CNN's accuracy. Sensitivity, Specificity, and Precision were employed alongside accuracy to comprehensively evaluate the network's performance. Notably, the SoftMax classifier demonstrated the highest accuracy among the categorizers, achieving 99.52% accuracy on test data. The presented research contributes to the growing field of deep learning in medical image analysis. The combination of CNNs and MRI data offers a promising tool for accurately detecting brain tumors, with potential implications for early diagnosis and improved patient care.

Motivation & Objective

• Motivate accurate early detection of brain tumors using MRI data.
• Evaluate CNN-based architectures with different classifiers for MRI tumor detection.
• Assess the impact of clustering-based feature extraction on CNN performance.

Proposed method

• Train CNNs on MRI scans of healthy and tumor patients.
• Classify with SoftMax Fully Connected layer and compare to RBF and Decision Tree classifiers.
• Introduce a clustering-based feature extraction method to enhance CNN accuracy.
• Evaluate performance using accuracy, sensitivity, specificity, and precision.
• Report that SoftMax achieves highest accuracy on test data.

Experimental results

Research questions

• RQ1 How do CNNs with different classifiers (SoftMax, RBF, DT) perform on brain tumor detection from MRI images?
• RQ2 Does clustering-based feature extraction improve CNN-based tumor classification?
• RQ3 Which classifier yields the best overall performance metrics (accuracy, sensitivity, specificity, precision) on test data?

Key findings

• SoftMax classifier achieves 98% accuracy on training/validation and 99.52% accuracy on test data.
• RBF classifier achieves 98.24% accuracy, while the Decision Tree achieves 95.64% accuracy on test data.
• Clustering-based feature extraction improves CNN accuracy over baseline CNN without clustering.
• The study uses sensitivity, specificity, and precision alongside accuracy for evaluation.