Automatic Brain Tumor Detection And Classification Using Modified Densenet201

Authors

  • Mukesh Kumar Tripathi, Sandeep Kadam, Chaitali Shewale, Veena Kadam, Amrin Sheikh, Sonali Shirke Author

DOI:

https://doi.org/10.64252/bnjbyt92

Keywords:

Brain tumor, MRI images, glioma, meningioma, pituitary, DenseNet201

Abstract

The brain is the most significant organ in the body since it is responsible for controlling all of the other organs. A tumor is an abnormal development of cells that results from the unregulated division of cells. This results in the formation of the tumor. There are three different sorts of tumors namely glioma, meningioma and pituitary. In a variety of fields, including medical imaging, DL-based algorithms have demonstrated exceptionally high levels of performance. In this article, an automated method for the identification and categorization of brain tumors by using DenseNet201 deep learning model. The conventional DenseNet201 model is modified by adding dropout layer to remove extra connections and make the model optimal. A Dense Net is a specific kind of deep learning model that makes use of dense connections between layers. These connections are made using Dense Blocks, which include connecting all layers directly with each other and ensuring that their feature-map sizes are the same. The proposed model obtained an accuracy of 97.45% which is better than the existing models.

Downloads

Download data is not yet available.

Downloads

Published

2025-07-02

Issue

Section

Articles

How to Cite

Automatic Brain Tumor Detection And Classification Using Modified Densenet201. (2025). International Journal of Environmental Sciences, 1927-1937. https://doi.org/10.64252/bnjbyt92