Application of Gist SVM in Cancer Detection
This work addresses cancer diagnosis for medical applications, but it is incremental as it builds on existing SVM methods with specific optimizations.
The paper tackled cancer detection by applying GIST SVM to differentiate between benign and malignant cells, achieving improved classification accuracy through optimization of training set size and feature selection.
In this paper, we study the application of GIST SVM in disease prediction (detection of cancer). Pattern classification problems can be effectively solved by Support vector machines. Here we propose a classifier which can differentiate patients having benign and malignant cancer cells. To improve the accuracy of classification, we propose to determine the optimal size of the training set and perform feature selection. To find the optimal size of the training set, different sizes of training sets are experimented and the one with highest classification rate is selected. The optimal features are selected through their F-Scores.