Lung Cancer Detection in CT Scans Employing Image Processing Techniques and Classification by Decision Tree(DT) and K-Nearest Neighbor(KNN)
Lakshmi Hanisha Jyothula, Greeshma Eppa, Abhishikth Varma Indhukuri, Mohd. Javeed Mehdi · 2023
Lung cancer is a hazardous, and also challenging cancer to diagnose. Since it frequently results in mortality so quickly, proper nodule analysis is particularly crucial for treatment. Initial noticing and prognosis help raise individual life expectancies. Low-dose CT is the only screening procedure that is advised for lung cancer. Screening is done to discover diseases early (LDCT). Finding aberrant lung tissue that may be cancerous with the aid of LDCT scans is possible. Images from CT scans, also known as computed tomography, are used to detect cancer. Although CT is recommended, visual interpretation of these images has the potential to be inaccurate and can delay the discovery of tumor. In order to detect lung cancer in the initial phases, Image Processing approaches are commonly used within the medical industry. An algorithmic approach for spotting melanoma in CT scan pictures is presented in this study. By employing techniques like median filter in Image Pre-processing and Morphological system for partition the lung ROI, Classification stage to categorize CT scan pictures into Normal or Cancer by computing shape based parameters such as Area, Compactness, Euler, Circularity, and Perimeter. Finally, the accuracy rates of Decision Tree(DT), K- Nearest Neighbor(KNN) are acquired as 93.88% and 94.44%, respectively.