BRAIN TUMOR CLASSIFICATION USING FCM AND TRANSFER LEARNING TECHNIQUES

Abstract

Brain tumor classification is a critical task in medical diagnostics, requiring high accuracy and precision for effective treatment planning. This study presents a novel approach that integrates Fuzzy C-Means (FCM) clustering and transfer learning techniques to enhance the accuracy of brain tumor classification. FCM is utilized for its ability to handle uncertainty and imprecision in image data, effectively segmenting the brain tumor regions. Subsequently, transfer learning is employed, leveraging pre-trained deep learning models to extract relevant features from the segmented images. The pre-trained models are fine-tuned on a specific brain tumor dataset, ensuring that the model adapts to the unique characteristics of the dataset. Our proposed method demonstrates superior performance in distinguishing between different types of brain tumors compared to traditional methods. The integration of FCM with transfer learning not only improves segmentation accuracy but also enhances feature extraction, leading to more accurate classifications. The experimental results indicate that this hybrid approach achieves higher accuracy, sensitivity, and specificity, making it a promising tool for aiding radiologists in the early detection and classification of brain tumors. This methodology could potentially reduce diagnostic errors and improve patient outcomes by facilitating timely and accurate diagnosis.