ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Enhancing brain tumor detection: integrating CNN-LSTM and CNN-BiLSTM models for efficient classification in MRI images

Journal: International Journal of Advanced Technology and Engineering Exploration (IJATEE) (Vol.11, No. 115)

Publication Date:

Authors : ; ;

Page : 888-898

Keywords : Machine learning; Deep learning; CNN; LSTM; BiLSTM; Tumor detection; FCM.;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

Brain tumors are among the leading causes of mortality in humans, characterized by their low survival rates due to the aggressive nature of these tumors. Accurate diagnosis of various malignant and benign brain tumors is crucial. Magnetic resonance imaging (MRI) provides detailed internal views of the human brain, aiding doctors and radiologists in diagnosing brain tumors. However, interpreting MRI images involves complex details that require extensive time and expertise. Artificial intelligence offers solutions to these challenges by simplifying the analysis process. This study aims to develop a fast and accurate system for brain tumor detection. The initial phase of the proposed system involves a segmentation process, where the tumor is distinguished from the background using the fuzzy c-means (FCM) algorithm, resulting in images segmented into foreground and background. These images are then input into the proposed convolutional neural network-long short-term memory (CNN-LSTM) and convolutional neural network-bidirectional long short-term memory (CNN-BiLSTM) models for feature extraction and tumor identification. The goal of this work is to enhance the performance of brain tumor classification and reduce training times. Experimental results demonstrate the effectiveness of the models. The LSTM classifier model was trained in 58 seconds, and the BiLSTM classifier in 91 seconds, achieving accuracies of 97.86% and 99.77%, respectively. However, one limitation noted was the small size of the dataset used in the experiments, which may affect the generalizability of the results.

Last modified: 2024-07-04 18:01:47