Advancements in Brain Tumor Detection: Utilizing Xception Enhanced Tumor Identifier Network

Sandeep Shiraskar, Dominick Rizk

Abstract: Detecting brain tumors accurately and efficiently is crucial due to the life-threatening nature of these conditions and the significant impact on patient outcomes. This paper presents a novel approach to brain tumor detection utilizing Convolutional Neural Networks (CNN) on a dataset of MRI brain tumor samples. By leveraging the XETINet (Xception Enhanced Tumor Identifier Network) network, our model achieves superior accuracy compared to existing models, underscoring the potential for improved diagnostic tools in medical imaging. The Xception architecture’s depthwise separable convolutions allow for enhanced feature extraction and model performance, making it particularly well-suited for this application. Our research demonstrates the importance of early and precise brain tumor detection in improving treatment efficacy and patient survival rates. The high accuracy of our model highlights its potential as a reliable diagnostic aid, capable of assisting radiologists in identifying tumors with greater confidence and speed. Recognizing the need for accessible and user-friendly diagnostic tools, we are also working on refining this algorithm to ensure ease of implementation. Our goal is to develop a model that not only surpasses existing public and published algorithms in accuracy but also facilitates widespread adoption in clinical settings. This paper contributes to the field of medical imaging by offering a state-of-the-art solution for brain tumor detection, emphasizing both technological advancement and practical applications. By addressing the critical need for effective diagnostic tools, our research aims to significantly impact the landscape of brain tumor diagnosis and treatment.

External IDs:dblp:conf/aibthings/ShiraskarR24