LoG Feature Extraction and Supervised SVM Classification for Accurate Lung Malignancy Detection and Classification in CT Scans

Pulmonary cancer is still one of the most prevalent forms of carcinoma along with breast cancer is the realms of growing focus for fundamentals healthcare. This project focuses on the identification of lung neoplasm based on the study of CT scan images using these methods. In this case investigation begins with image preparation, which Median filtering is used to decrease noise. while maintaining important image data. mathematical morphological transformations are then used for picture segmentation. The goal is to separate tumor tissues from the remainder of the image for additional investigation. Area, perimeter, eccentricity and texture features from the segmented images are extracted for GLCM, which are then supplied to the SVM classifier. This task is best suited for the SVM because it is known that the SVM is good at working within high dimensional spaces and is proficient at separating cancerous from non cancerous tumors. First estimates prove that the proposed system is a lot better than traditional systems at pose estimation detection because there are less mistakes made by the operator. The incorporation of these advanced image processing techniques together with machine learning algorithms have great prospects for boosting the detection rate of lung cancer.