Comparison between fuzzy kernel c-means, fuzzy kernel possibilistic c-means and support vector machines in soft tissue tumor classification

Abstract

Soft Tissue Tumor (STT) are cell growths, whose existence are not limited to the presence of tumors in soft tissues. Furthermore, they are classified into soft tissue and non-soft tissue tumor and early detection is important to determine the right course of treatment. This research, therefore, aims to compare fuzzy kernel c-means, fuzzy kernel possibilistic c-means and support vector machines on Soft Tissue Tumor dataset, obtained from Nur Hidayah Hospital, Yogyakarta, Indonesia, consisting of 50 STT and 25 non-STT samples. The results conclude that fuzzy kernel c-means provides a better running time when using the parameter σ = 0.05. However, support vector machines, with the parameter σ = 0.0001 performs better than other methods in terms of accuracy, sensitivity, precision, and F1-Score.