Akademik Veri Yönetim Sistemi
4th International Conference on Intelligent and Fuzzy Systems (INFUS), Bornova, Türkiye, 19 - 21 Temmuz 2022, cilt.505, ss.463-471
Lung cancer is one of the most common causes of deaths worldwide. The ability to predict and diagnose cancer has become increasingly significant in recent years. Early identification of lung cancer appears to be the only way to improve patients' survival rates, which is a difficult effort due to the structure of cancer cells, which has most of the cells overlapping each other. For lung cancer prediction and early diagnosis, image processing techniques are commonly used. The goal of this work is to improve lung CT scans so that the tumour may be identified quickly. First, certain enhancement approaches are employed to enhance and remove noise in photos during image pre-processing. Different sections of the photographs are isolated in the next stage, and the tumour is segmented in later stages. The Improved Linear Iterative Clustering methodology (ILIC) is offered as the first step in our suggested method, whereas the existing system used Simple Linear Iterative Clustering (SLIC), which has various shortcomings. The Improved Linear Iterative Clustering (ILIC) technique is widely used in various types of image processing because of its perceptually excellent meaningful qualities. The proposed technique is applied, and it has improved boundary recall, fuzzy boundary robustness, super pixel size setting, and preprocessing performance on lung photographs. Finally, the PSNR, MSE, and SSIM methods are derived in the preprocessing section.