A NOVEL ROBUST HYBRID FEATURE ION ALGORITHM FOR OPTIMAL CLASSIFICATION OF SINUSITIS DETECTION USING NEURAL NETWORK

A novel hybrid feature selection algorithm combining Spotted Hyena Optimization (SHO) and Rain Optimization Algorithm (ROA) is proposed in this work for supporting the optimal classification of sinusitis.The proposed Deep Learning (DL) based detection model uses Adaptive Gaussian Notch Filter (AGNF) for preprocessing the input CT scans of paranasal sinus.Using adaptive window size variation, the AGNF significantly defuses the noisy peaks in the input image.Then with the application of Recurrent Neural Network (RNN), the relevant Region of Interests (ROIs) that buttress the accurate sinusitis classification are segmented separately.The critical features including homogeneity, contrast, ASM, energy, dissimilarity and homogeneity are extracted from the segmented image using Gray-Level Co-occurrence Matrix (GLCM).Then using the proposed robust Hybrid Rain Optimized Spotted Hyena Optimizer (HRSHO), the best feature subset that improves the performance of Convolution Neural Network (CNN) classifier is selected.In the proposed HRSHO, the exploration and exploitation phase of the SHO is improved further using ROA.The proposed DL based sinusitis is examined for its working by executing in Python software.Moreover, from the derived results it is noted that the proposed HRSHO operates as an excellent feature selection technique, as it selects minimum number of features by excluding the irrelevant redundant features.Additionally, it is also observed that the HRSHO has a significantly higher accuracy and minimum computational cost.