A Comprehensive Survey on Lung Cancer Disease Detection

This work automates the diagnosis of lung cancer from CT scan pictures using image processing techniques. Raising survival rates for lung cancer, a prevalent and deadly illness, hinges on earliest detection. Computed tomography (CT) scans can identify lung cancer, but their visual interpretation can be difficult and prone to mistakes. So, the method suggested in this paper involves preprocessing the image, segmenting the lung region of interest, feature extraction, and classification using a Support Vector Machine (SVM). Median filtering for noise reduction and contrast enhancement for better image quality are both pre-processing techniques. Accurate lung region segmentation is accomplished using mathematical morphological methods. The segmented tumor region is used to extract three geometric features : perimeter, eccentricity, and area. SVM is used to classify CT scan images into normal and pathological categories based on these attributes. The empirical findings show how well the suggested method works for correctly identifying lung cancer in CT scans. This automated technology could help radiologists identify patients more quickly and accurately, ultimately leading to better patient outcomes.