Hybrid Radial Basis Function and Support Vector Machine Model for Precise Breast Cancer Diagnosis
Rishabh Sharma, Praney Madan, Shanmugasundaram Hariharan, Shubham Mahajan · 2024
The study is conducted to add a radial basis function (RBF) into the Support Vector Machine (SVM) detection mechanism to increase the accuracy of mammographic image recognition. We resorted to the mammographic images dataset where the data was prepared, augmented, and pre-trained to enhance the model performance. Feature extraction was performed to derive the principal image features and then the feature selection was done to reduce the noisy elements of the dataset to optimize the SVM performance. The SVM model was set up with RBF kernel and also it used Grid search and cross-validation techniques for optimization of gamma and cost parameters to determine the best settings. The integrated RBF-SVM model was better at classifying RBF-SVM compared to models using linear and polynomial kernels. We've got the sensors placed correctly, then the model has 92.22% accuracy, 89.93% precision, 94.52% recall, and 91.5% F1-score. The statistical analysis indeed indicated that the uplift of the model performance parameters was significant with p values minor than 0. 05 or 0. 01. 05. The RBF kernel which was incorporated with the SVM classifier resulted in a better detection of breast cancer in the mammogram class. It is this model with the paramount recall rate (greater than for the other algorithms) that helps clinics diagnose a smaller proportion of malignant cases. The research allows us to conclude that the ML models, particularly the ones with the RBF kernel, are very effective in the process of breast cancer screening as they could highly improve the accuracy and reliability of cancer detection mass screening. The research not only indicates the opportunities for complex AI-based techniques but also points to their usefulness in diagnostic medical applications. Next research should involve verification of the model in different contexts and also seeing to it that the model is implemented into clinical practice and thus ensures it is effective .