Utilizing Deep Learning Techniques to Diagnose Nodules in Lung Computed Tomography (CT) Scan Images

There are different methods available for detecting lung cancer including CT (Computed Tomography) scan, MRI(Magnetic Resonance Imaging) scan etc. Among all methods, CT scan images are preferred more because they can detect a very small nodule in the lungs. Early treatment can be given to patients if it is diagnosed at early stages, hence reducing the number of deaths. This paper shows that Median Filter outperformed the Average, Gaussian, Laplacian, and Wiener Filters in the preprocessing stage for the removal of noise from images. Additionally, a study has been conducted on several image segmentation algorithms, such as clustering, watershed, and Thresholding segmentation. This was followed by the extraction and classification of nodules. Different performance parameters have been calculated to validate the results of the model and it is discovered that proposed model has greatest performance.