Research on Feature Selection and Classification Algorithm of Medical Optical Tomography Images

• Published in: 2019 IEEE/CIC International Conference on Communications in China (ICCC) pp. 561 - 566
• Main Authors: Ji, Hongkun, Guo, Kaiwen, Yang, Lizhu, Jiang, Weinan, Zhao, Zijun, Zhu, Xukang, Hou, Alin
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2019
• Subjects: Classification algorithms; cluster analysis; Feature extraction; feature selection; Medical diagnostic imaging; medical image; Particle swarm optimization; Support vector machines; US Department of Transportation
• DOI: 10.1109/ICCChina.2019.8855804

With the continuous development of the computer vision technology and medical imaging equipment, the information contained in medical images is extremely rich. Therefore, as the key point of image classification, the extraction and selection of medical image features has become more and more important. The dimension of the image information is getting higher and higher, so it is very important to select more effective features to classify the images. In this paper, the medical images of rheumatoid arthritis (RA) were studied, and the optical coefficients were extracted as the feature from diffusion optical tomography (DOT) of the fingers of RA patients. Feature selection is the key method of medical image diagnosis. Consequently, the classification of RA medical images is studied by using the maximum relevance and minimal redundancy (MRMR) feature selection algorithm with the weight value improvement. In order to improve the accuracy rate of the classification, particle swarm optimization (PSO) is used to optimize the parameters of support vector machine (SVM) before the medical image classification. The simulation results show that the method is effective, and has the reference value for the medical image processing which contains numerous features and needs reasonable selection.