Abstract:
The leading cause of cancer-related fatalities globally is lung cancer. Your
airways, lung tissues, or blood that flow into and out of your lungs may all be affected
by lung disorders. Early detection of lung disorders is essential, especially after the wake
of the COVID19 epidemic. Therefore, the survival rate of patients is significantly
influenced by early therapy. A system that can help radiologists identify CT scans is
required to diagnose CT scans more quickly and to minimize human error. In this work,
we compared three segmentation techniques: K-means clustering, Fuzzy C-means
(FCM), as baseline methods, and Superpixel-based Fast Fuzzy C-means (SFFCM) as the
proposed method, using synthetic images and lung CT scan images. Results showed that
the proposed method (SFFCM) yields better segmentation results and has more
robustness than the two baseline methods (K-means and FCM).
Keywords: Image Segmentation, Computed Tomography (CT) Images, Superpixel-based Fast
Fuzzy C-means, K-means, Fuzzy C-means, Lung Cancer Detection
