Machine Learning Models and Innovative Feature Design for the Classification of Lung Cancer

Lung cancer accounts for the majority of cancer-related deaths worldwide because of its high incidence and usually delayed diagnosis. It affects public health significantly and results in noteworthy rates of morbidity and mortality. A patients quality of life and chance of survival depend on prompt diagnosis and effective care. Currently computer vision and image processing methods are very beneficial for classifying lung cancer. By combining machine learning and manually created features the model presented in this paper effectively classifies lung cancer from CT scan images. The model begins by applying Gaussian filtering (GF) to preprocess and improve the quality of the input images. After that an image slice segmentation technique is used to accurately identify the diseased areas of the images. The Oriented Rapid and Rotated Abstraction (ORB) and Gray Level Co-occurrence Matrix (GLCM) methods are used to extract features. Finally using a Random Forest (RF) classifier the right classifiers for experimental lung cancer images are found. The efficacy and efficiency of the proposed method are evaluated against other current methods using a dataset of CT images of lung cancer. The recommended model showed excellent accuracy and efficiency scoring a 95. twenty-three percent. The experimental results which show the models superiority in a number of performance metrics illustrate its potential for practical application in lung cancer diagnosis and treatment planning.