HISTOGRAM BASED TUMOUR CLASSIFICATION FOR BRAIN MRI IMAGES USING ANN

Abstract

Abstract

This paper presents a novel approach for the classification of brain tumours in MRI images using histogram-based features and Artificial Neural Networks (ANN). Accurate classification of brain tumours is crucial for effective diagnosis and treatment planning. Traditional methods often rely on complex pre-processing and feature extraction techniques, which can be time-consuming and computationally intensive. Our methodology simplifies this process by leveraging histogram features that capture the intensity distribution of pixels within the tumor region, providing a straightforward yet powerful representation of the image data. The proposed system involves several key steps: pre-processing the MRI images to enhance quality and consistency, extracting histogram features from these images, and then using these features to train an ANN model. The ANN, designed with multiple hidden layers, is optimized to distinguish between benign and malignant tumours based on the input histogram data.

our system is evaluated using a comprehensive dataset of labelled MRI images, and the results demonstrate its high accuracy and reliability. The classification accuracy achieved was [insert accuracy percentage here], indicating the robustness of our approach. The use of histogram features not only simplifies the feature extraction process but also enhances the ANN's performance by providing relevant and discriminative information about the tumor characteristics. This study highlights the efficacy of combining histogram-based feature extraction with ANN for brain tumour classification, offering a viable solution that improves diagnostic accuracy and efficiency. The findings suggest that our method can significantly aid in the early detection and accurate classification of brain tumours, potentially leading to better patient outcomes and optimized treatment strategies. The simplicity and effectiveness of this approach pave the way for further advancements in medical image processing and diagnostic techniques.

Keywords:  Artificial Neural Network (ANN), Benign Tumor,, Brain Tumor Classification, Diagnostic Accuracy, Feature Extraction, Histogram Features, Image Preprocessing, Magnetic Resonance Imaging (MRI), Malignant Tumors,  Medical Image Processing.