Brain Tumor Detection Using MobileNetV2 and Linear K-Means Support Vector Machine (LK-SVM)

Abstract

This paper presents an innovative approach for brain tumor detection by leveraging the capabilities of MobileNetV2 and Linear K-Means Support Vector Machine (LK-SVM). The proposed method combines the efficiency of MobileNetV2's deep learning architecture with the precision of LK-SVM for accurate and robust tumor classification. MobileNetV2 is utilized for feature extraction due to its lightweight structure and high performance in capturing intricate patterns within medical imaging data. Subsequently, the extracted features are fed into the LK-SVM for classification, enhancing the overall detection accuracy. This hybrid model is evaluated on a comprehensive dataset of brain MRI images, demonstrating superior performance in terms of accuracy, sensitivity, and specificity compared to traditional methods. The results indicate that the MobileNetV2 and LK-SVM combination can serve as a powerful tool in the early diagnosis and treatment planning of brain tumors, potentially improving patient outcomes through timely and accurate detection.