Non Small Cell Lung Cancer Detection and Stage Classification Using Modified CNN

Abstract

Lung cancer is a deadly illness with the highest rates of infection and mortality of any cancer in the world. Considering effective therapy options are heavily dependent on the individual stage of cancer, pattern classification of lung cancer can considerably lower mortality rates. Unfortunately, due to time constraints and additional costs, manual staging remains a barrier. Various researchers started designing a lung detection module based on the mathematical and meta-heuristics used to address this. The limitations of the existing detection techniques are (i) lower accuracy rate, (ii) lung stage classification has not been focused much on, and (iii) CAD assist decision-making automatic framework which detects cancer at its earlier stage has not been considered in the previous lung tumor identification. The authors concentrated on either segmentation or detection of lung tumors, but not both. In this research, we considered quotlung tumor dissection, detection of nodules, and multi-stage classificationquot together and presented the efficient results and the CAD assist system newline