Paper Contents
Abstract
This project proposes detection of MRI-based brain tumor classification through the utilization of MobileNet and Streamlit. Initially, the input image dataset is allowed to undergo a preprocessing stage to enhance the quality of the image. This stage is handled by a Gaussian filter. Then, in the next stage, the FCM segmentation technique is used to accurately determine the variety in size, location, and shape of tumors. Finally, the extraction and classification of features are implemented using the MobileNet framework. Mobilenet is one of the methods in deep learning that performs the segmentation process of medical images. Mobile Network is a CNN model with high accuracy and less computation. In regards to precision, recall, and accuracy, the F-score is calculated using a performance matrix. This work is developed using the Streamlit environment, an open-source Python platform for developing web-based interactive applications. Finally, the predicted output image for types of glioma, meningioma, pituitary, and no-tumor is displayed in the Streamlit software. This work provides a quick and accurate approach to classifying diseases, which helps the doctor detect patients suffering from tumors, and the entire process reduces the computation time. This project is implemented using Python Jupyter software and the Python programming language.
Copyright
Copyright © 2025 N.DEENA NEPOLIAN. This is an open access article distributed under the Creative Commons Attribution License.
