Clarke L P, Velthuizen R P, Phuphanich S, Schellenberg J D, Arrington J A, Silbiger M
Center for Engineering and Medical Image Analysis (CEMIA), College of Engineering, University of South Florida, Tampa 33612.
Magn Reson Imaging. 1993;11(1):95-106. doi: 10.1016/0730-725x(93)90417-c.
Supervised segmentation methods from three families of pattern recognition techniques were used to segment multispectral MRI data. Studied were the maximum likelihood method (MLM), k-nearest neighbors (k-NN), and a back-propagation artificial neural net (ANN). Performance was measured in terms of execution speed, and stability for the selection of training data, namely, region of interest (ROI) selection, and interslice and interpatient classifications. MLM proved to have the smallest execution times, but demonstrated the least stability. k-NN showed the best stability for training data selection. To evaluate the segmentation techniques, multispectral images were used of normal volunteers and patients with gliomas, the latter with and without MR contrast material. All measures applied indicated that k-NN provides the best results.
我们使用了来自三个模式识别技术家族的监督分割方法对多光谱MRI数据进行分割。研究对象包括最大似然法（MLM）、k近邻法（k-NN）和反向传播人工神经网络（ANN）。我们从执行速度以及训练数据选择（即感兴趣区域（ROI）选择、层间和患者间分类）的稳定性方面对性能进行了评估。结果表明，MLM的执行时间最短，但稳定性最差。k-NN在训练数据选择方面表现出最佳的稳定性。为了评估分割技术，我们使用了正常志愿者和患有神经胶质瘤患者的多光谱图像，后者使用了和未使用磁共振造影剂。所有应用的测量方法均表明，k-NN提供了最佳结果。
