Enhanced image registration based brain tumour segmentation using optical particle swarm intelligence technique with Resnet Inceptionv2 HCNN.

A brain tumor is the deadliest disease to cause sudden death, affecting billions of people worldwide. Artificial Intelligence (AI) powered technologies play a vital role in screening medical images to identify brain-suspecting tissue regions of attained diseases for early diagnosis and treatment to avoid brain disorders. In the preliminary stages, the Machine Learning (ML) and Deep Learning (DL) models have potential impacts on identifying brain tumors and disorders. Due to increasing feature dissimilarities, the problem arises due to image entity registration and improper segmentation, leading to poor precision identification accuracy due to higher concurrent false negatives. To solve these problems, propose an optimal image feature engineering-based Optical Particle Swarm Intelligence Technique (OPSIT) for feature selection with Resnet-incptionv2-Hyper Convolution Neural Network to identify the disease effectively. Initially, the preprocessing is carried out by Cascaded Absolute Median Filter (CAMF) with an Otsu threshold margin to normalize the feature scaling and improve the image scalar margins. Then, Active Contour Colour Histogram Evolution (ACCHE) is applied to vary the tumor region, and Hyperactive slicing window segmentation is used to split the extractive feature of the disease region. Further, to scale the active disease margins in the tumor region, the ideal features are evaluated with OPSIT to reduce the non-relation feature in the segmented image region. Finally, the Resnet inceptionv2-HCNN algorithm is applied to train the scaled entity of the brain tumor region, with Actual threshold margins to identify the disease region accurately. The proposed system increases the proper positive actual scaling region of the tumor detection region to increase the precision rate and attain high accuracy, sensitivity, specificity, and ROC performance compared to the other systems.