Independent Weighted Feature Set with Linked Feature Reduction Model for Lung Cancer Stage Detection using Machine Learning Model

Abstract

Lung cancer is a potentially fatal disease that is affected to 18% of population every year. Finding the exact location of a cancer and identification of lung cancer stage continues to be difficult for medical professionals. The true reason for cancer and a comprehensive cure is still unknown. Treatment for cancer is possible if detected at an early stage with accurate stage detection. Finding areas of the lung that have been impacted by cancer requires the use of image processing techniques like noise reduction, highlight filtration, recognizable proof of effected lung regions, and perhaps a comparison with data on the curative history of lung cancer. This research investigates whether or not technology enabled by machine learning algorithms and image processing can correctly classifies and predict lung cancer. For images, the dimensional feature channel is used in the preliminary processing stage. The proposed model considers Magnetic Resonance Imaging (MRI) images for detection of lung cancer. This research proposes an Independent Weighted Feature Set with Linked Feature Reduction (IWFS-LFR) model for accurate lung cancer stage detection based on the size of the tumour. The tumour stage can be accurately predicted using the feature attribute similarity calculation for accurate detection of lung cancer stage for proper diagnosis. The proposed model when contrasted with the traditional model exhibits better performance.