Treffer: Ensemble learning for enhanced brain tumor diagnosis: A new approach for early detection.

Brain tumors represent one of the most extreme and complex types of cancer, requiring unique analysis for powerful remedy and management. Accurate and early identification of brain tumors can greatly enhance patient outcomes and decrease mortality. Nowadays, deep learning aids the medical field a lot by diagnosing magnetic resonance imaging images in brain tumors. The potential of deep learning architectures to improve brain tumor diagnosis accuracy was explored in this work. This study evaluated three different convolutional neural network architectures: AlexNet, VGG16, and ResNet18 as an ensemble model. By leveraging the complementary strengths of these models and applying them to a dataset sourced from local hospitals and public repositories, this research aims to address the challenges in accurate and early brain tumor detection. Our ensemble technique achieved excessive accuracy, demonstrating its potential for reliable computer-aided diagnosis (CAD) in medical imaging. However, while the results indicate an improvement in class overall performance, the novelty of this approach is restrained because it builds upon existing methodologies as opposed to offering a completely new framework. The gathered dataset was used to train and test the models. To enhance the dataset’s balance and the models’ performance, data were collected from Rizgary Hospital (Erbil) and Hiwa Hospital (Slemani), addressing the underrepresentation of cases from the Kurdistan Region of Iraq (KRI). These image enhancement techniques were applied to two categories: normal and abnormal brain tumors. Several brain tumor datasets are available online for the development of CADs, but not KRI cases, which pose challenges in their classification through deep learning models. This study was implemented with Python programming language. Out of the three models, ResNet had the highest accuracy of 98.66%, VGG16 had an accuracy of 97.8%, and AlexNet had an accuracy rate of 97.666%. The ensemble, using both majority voting and weighting voting strategies, achieved an accuracy of 98.33%. [ABSTRACT FROM AUTHOR]

Copyright of International Journal of Systematic Innovation is the property of Editorial Office of International Journal of Systematic Innovation and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)