Multilevel Optimization with Hybrid Stack Model for Lung Cancer Classification

Abstract

Lung cancer is the leading cause of cancer-related mortality worldwide, and detecting the disease can still save lives. if a lung cancer diagnosis has been made. The illness known as lung cancer occurs when healthy lung cells transform into dangerous aberrant cells known as cancer cells. Tumors are collections of cancerous cells that grow over time. Today, medical imaging scans are being interpreted with the use of artificial intelligence and contemporary data science techniques. In contrast to conventional techniques, which depend on the subjective and time-consuming visual examination of radiologists, the emphasis now is on creating reliable automated diagnostic tools. This change is in line with the core objective of radiomics, a developing field of study that combines customized medicine with medical imaging. Improvements in lung cancer screening provide a glimmer of hope for life-saving treatments. Together, in stage one, the survival rate is 70%. Stage 2 drops to 50% as stage 3 drops. And stage four for the most part is not curable disease, but there are some patients who might be alive around five years.The aforementioned problems are efficiently addressed by the merging of LDCT followed by AI based Multilevel Optimization with hybrid stack model. This paper introduces a hierarchical reference architecture for Lung Cancer Classification. The proposed approach (GLCM) features are extracted from LDCT further investigated with M-GWO technique to figure-out best features. The best solution obtained from this hybrid stacked model is use to classify input image as normal or abnormal.