An Experimental Model to Predict Bone Cancer Disease by Using Hybrid Learning Principles with Image Processing Technique

Bone cancer is a rare but aggressive disease that requires early and accurate detection to ensure better outcomes for patients. This paper presents a novel approach for predicting bone cancer disease using hybrid learning principles integrated with advanced image processing techniques. The proposed model combines Convolutional Neural Networks (CNN) for feature extraction with Random Forest for classification, utilizing AutoEncoders for dimensionality reduction to optimize performance. The methodology is applied to a comprehensive bone cancer dataset consisting of X-ray images collected from various public and medical sources. A robust preprocessing pipeline, including noise reduction, resizing, and segmentation, enhances image quality for precise feature extraction. The model was compared against nine existing models, including CNN, SVM, DenseNet, and VGG16. Performance evaluation using metrics such as accuracy, precision, recall, F1-Score, and AUC demonstrates the superior performance of the proposed model. The model achieved an accuracy of 97.81%, significantly outperforming existing models like VGG16 and DenseNet. The hybrid model's ability to minimize both false positives and false negatives makes it highly effective for medical diagnosis applications.