Applying the Particle Swarm Optimization and Boltzmann Function for Feature Selection and Classification of Lymph Node in Ultrasound Images

A lymph node (LN), which can resist virus and germs, is part of the lymphatic system that exists in the human body and every apparatus inside it. There are many kinds of pathological changes in LN. Metastatic is one of the important indexes to estimate the stage of malignant tumors. One convenient tool to observe LN is the use olf ultrasonic images. Clinical physicians judge a nosology by biopsy and experience. Shortcoming of the method is that it requires lots of precious time of clinical physicians. In this paper, we propose a method that classifies lymph node with different pathological changes in ultrasonic images. Features are extracted and selected from ultrasonic images. A feature selection method, which integrates the particle swarm optimization neural network (PSONN) with Boltzmann probabilistic, is proposed. Then, a support vector machine (SVM) is adopted for Lymph node classification. Experimental results show that the proposed approach decreases the number of the selected features and achieves a high accuracy in classification.