A Novel Assessment of Lung Cancer Classification System Using Binary Grasshopper with Artificial Bee Optimisation Algorithm with Double Deep Neural Network Classifier

Lung cancer is the leading cause of death from cancer worldwide. Finding pulmonary nodules is a critical step in the diagnosis of early-stage lung cancer. It has the potential to become a tumour. Computed tomography (CT) scans for lung disease analysis offer useful data. Finding malignant pulmonary nodules and determining whether lung cancer is benign or malignant are the main goals. Before further image preparation, image denoising is an important process for removing noise from images (feature extraction, segmentation, surface detection, and so on) maximising the preservation of edges and other intact features. This study employs a novel evolutionary method dubbed the binary grasshopper optimisation algorithm in order to address some of the shortcomings of feature selection and provide an efficient feature selection algorithm, the artificial bee colony (BGOA-ABC) algorithm enhance categorisation. Then, to categorise the chosen features, we employ a hybrid classifier known as a double deep neural network (DDNN) algorithm. A technique used by MATLAB that segments impacted areas utilising improved IPCT (Profuse) aggregation technology employing datasets from the cancer image archive (CIA), the image database resource initiative (IDRI), and the lung imaging database consortium. Various performance metrics are evaluated and associated with different cutting-edge techniques, classifiers that are already in use.