Kyriacou Efthyvoulos, Pattichis Marios S, Christodoulou Christodoulos I, Pattichis Constantinos S, Kakkos Stavros, Griffin Maura, Nicolaides Andrew
Department of Computer Science, University of Cyprus, Nicosia, Cyprus.
Stud Health Technol Inform. 2005;113:241-75.
The aim of this chapter is to summarise the recent advances in ultrasonic plaque characterisation and to evaluate the efficacy of computer aided diagnosis based on neural and statistical classifiers using as input texture and morphological features. Several classifiers like the K-Nearest Neighbour (KNN) the Probabilistic Neural Network (PNN) and the Support Vecton Machine (SVM) are evaluated resulting to a diagnostic accuracy up to 71.2%.
本章的目的是总结超声菌斑特征分析的最新进展,并评估基于神经和统计分类器、使用纹理和形态特征作为输入的计算机辅助诊断的有效性。对几种分类器进行了评估,如K近邻(KNN)、概率神经网络(PNN)和支持向量机(SVM),诊断准确率高达71.2%。