An Efficient SVM Based Lymph Node Classification Approach Using Intelligent Communication Ant Colony Optimization

摘要

Ultrasound imaging employs high-frequency sound waves to visualize the interior portion of the human body. Since, the ultrasound images are acquired in real-time; it displays the motion of the interior organs along with the flow of blood via the circulatory system. Presently, in the clinical diagnosis of diseases and disorders, the ultrasound imaging is extensively deployed. Inopportunely, the ultrasound imaging system is considered to be unstable due to the fact that various kinds of system have its own distinct properties. A universal feature selection approach that utilizes the core concept in the ant colony algorithm for solving the problem in diverse ultrasound imaging systems is proposed in this work. Support vector machine mechanism is deployed for classifying the various kinds of lymph nodes based on the selected vital features. The experiments illustrate that the proposed intelligent communication ant colony optimization (ICACO) method provides superior lymph nodes classification results, when compared with existing methods.