International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 03 | Mar 2024
p-ISSN: 2395-0072
www.irjet.net
Brain Tumor Detection Using CNN Varun1, Rishabh Saklani2, Suyash Saini3, Dr. Kumod Kumar Gupta4 Student, Dept of CSE-AI, NIET Greater Noida, Uttar Pradesh, India, Student, Dept of CSE-AI, NIETGreater Noida, Uttar Pradesh, India, Student, Dept of CSE-AI, NIET Greater Noida, Uttar Pradesh, India, Professor, Dept Of CSE-AI, NIET Greater Noida, Uttar Pradesh, India. -----------------------------------------------------------------------------------***------------------------------------------------------------------------------------------
Abstract: This study introduces an innovative method for timely brain Tumor detection using a Convolutional Neural
Network (CNN) architecture, specifically employing the VGG16 model for feature extraction and transfer learning. Given the critical importance of early diagnosis, traditional manual image interpretation methods are replaced by deep learning techniques, which have shown promise in automating medical image analysis tasks. By leveraging the hierarchical representations learned by VGG16 on extensive image datasets, the proposed approach enhances detection accuracy and robustness. Evaluation on a benchmark dataset of MRI scans demonstrates the superiority of the CNN model with VGG16 over traditional machine learning methods and other deep learning architectures. Performance metrics such as accuracy, sensitivity, specificity, and AUC-ROC validate the effectiveness of the proposed method. Overall, this research offers a reliable and efficient solution for automated brain Tumors diagnosis, potentially revolutionizing clinical decision-making and patient management. By seamlessly integrating advanced technology with medical imaging, it addresses the critical need for early intervention and improved patient outcomes.
Keywords: Brain Tumor, Convolutional Neural Network, VGG16, Medical Imaging, Deep Learning, Diagnosis, Magnetic Resonance Imaging (MRI) 1.Introduction Brain Tumors pose a significant threat to human health, with both benign and malignant forms affecting millions worldwide. Timely detection is critical, and recent advancements in medical technology offer promising solutions through Artificial Intelligence (AI) and Machine Learning (ML)[1]. Utilizing sophisticated algorithms like Convolutional Neural Networks (CNNs) and the VGG16 model, AI-powered software can accurately detect and classify Tumors from MRI scans. Traditional methods of Tumors detection rely on manual interpretation of MRI images, a process that is labour-intensive and subjective [2]. However, the emergence of deep learning techniques, particularly CNNs, has revolutionized medical image analysis, providing automated and precise Tumors identification. Transfer learning further enhances accuracy by adapting pre-trained models like VGG16 to the specifics of brain MRI data. Automated Tumors detection systems are essential given the severity of brain Tumors and the limitations of manual interpretation. Magnetic Resonance Imaging (MRI) remains the primary diagnostic tool due to its ability to provide detailed images without radiation exposure [3]. Early detection significantly impacts patient survival rates, underscoring the importance of advanced imaging techniques in medical practice. Brain Tumors encompass a diverse range of abnormalities, both benign and malignant, impacting various aspects of human health. It is crucial to detect these Tumors early, as they can lead to severe consequences if left untreated [4]. Recent advancements in medical technology, particularly in the realm of AI and ML, offer promising avenues for early detection and classification of brain Tumors. By harnessing the power of advanced algorithms such as Convolutional Neural Networks (CNNs) and the VGG16 model, AI-driven software can analyse MRI scans with remarkable accuracy and efficiency [6]. These systems automate the detection process, reducing the reliance on manual interpretation by radiologists, which can be time-consuming and prone to errors. Transfer learning further enhances the performance of these algorithms by fine-tuning pre-trained models to the nuances of brain MRI data. This adaptation process ensures that the AI systems can effectively identify and classify Tumors, distinguishing between benign and malignant forms [7].
2. Literature Review Brain Excrescences are a critical medical condition that requires early and accurate discovery for effective treatment and bettered patient issues. Traditional styles of brain excrescence discovery, similar as glamorous resonance imaging (MRI) and reckoned tomography (CT) reviews, calculate on homemade interpretation by radiologists, which can be timeconsuming and prone to mortal error [8]. Accordingly, there's a growing need for automated and intelligent systems that can help in the discovery and bracket of brain excrescences with high delicacy and effectiveness. In recent times, deep literacy ways, particularly convolutional neural networks (CNNs), have demonstrated remarkable success in colourful
© 2024, IRJET
|
Impact Factor value: 8.226
|
ISO 9001:2008 Certified Journal
|
Page 1412