Strategic Integration of CNN, SVM, and XGBoost for Early-stage Tumor Detection using Hybrid Deep Learning Method

A tumor is an unnatural cell development that can develop into cancer. Most often, cancer affects the brain, kidneys, lungs, liver, and other organs of the body. MRI (Magnetic Resonance Imaging) scans are the official method for locating tumors. Detecting a tumor in its early stages and using the resulting image to spot aberrant tissue growth allows for early cancer treatment. Early detection of tumors plays a pivotal role in improving the prognosis and overall survival rates of patients. A Hybrid Model was constructed which composed of a self-defined Convolution Neural Network (CNN), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost) is applied in detecting the presence of tumors and their performance is analyzed and the Proposed system detects the Tumor in the beginning stage with the accuracy of 98%. The proposed approach leverages the strengths of each individual technique to address the challenges associated with tumor detection. CNNs are employed to extract intricate spatial features from medical imaging data, capturing relevant patterns and representations. SVM is then applied to reduce the dimensionality of the feature space, enabling efficient and effective classification. Finally, XGBoost is used to optimize the classification performance by leveraging ensemble learning and gradient boosting techniques. Experimental results on a diverse dataset demonstrate the superiority of the integrated approach, achieving state-of- the-art accuracy and outperforming traditional methods for early-stage tumor detection.