CAD Detection of Brain Tumors Using Convolutional Neural Networks (CNN)

Jessica Lee 1, * and Brian Thompson 2

1 Department of Engineering and Applied Sciences, Harvard University, USA.
2 Department of ECE, University of Illinois Urbana-Champaign, USA.
 
Research Article
World Journal of Advanced Engineering Technology and Sciences, 2024, 13(02), 821-834.
Article DOI: 10.30574/wjaets.2024.13.2.0540
Publication history: 
Received on 07 October 2024; revised on 22 November 2024; accepted on 25 November 2024
 
Abstract: 
This study is centered on using Convolutional Neural Networks (CNNs) to develop Computer-Aided Diagnosis (CAD) systems specifically for brain tumor detection. Given the persistent global health challenges posed by brain tumors, early and precise diagnosis is a key factor in improving patient outcomes. Traditional diagnostic methods heavily rely on radiology expertise, which is both subjective and time-consuming. In contrast, CNNs offer a robust deep learning solution capable of learning intricate features from medical images, particularly Magnetic Resonance Imaging (MRI), with high precision and efficiency.
In this study, we examine the effectiveness of different CNN architectures, from custom-designed to pre-trained models, for detecting and classifying various types of brain tumors. Major parts of the methodology are dataset choice (e.g., BraTS), preprocessing methods such as normalization and augmentation, and a solid training-validation-testing pipeline. Yardsticks measure performance, such as accuracy, precision, recall, F1-score, and AUC-ROC. Also, tools used to visualize model behaviors, such as Grad-CAM, explain model predictions and outline tumor regions, increasing model transparency.
The results of this study underscore the potential of CNN-based CAD systems to significantly enhance diagnostic speed and accuracy, making them a valuable resource for clinical settings. The study also addresses other challenges, such as data scarcity, generalization over imaging systems, and interpretability. The paper concludes with a discussion, suggesting future work that includes multi-modal data integration and the incorporation of Explainable AI (XAI) to boost clinical confidence and decision-making. This study highlights the promising prospect of advanced brain tumor diagnosis using CNNs through intelligent automated image analysis.
 
Keywords: 
Brain Tumor Detection; Convolutional Neural Networks; Computer-Aided Diagnosis; MRI Imaging; Deep Learning; Medical Image Analysis; Automated Diagnosis; Radiology; Feature Extraction; CNN Architecture; Transfer Learning; Diagnostic Accuracy; Neuroimaging; Clinical Decision Support; Artificial Intelligence in Healthcare
 
Full text article in PDF: 

Click here

 
Copyright information: 
Copyright © 2024 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0