图像分割是多维超声医学图像重建中最重要和最困难的问题。文中将传统的最近邻分类方法与自组织神经网络相结合，提出了一种超声医学图像的自动分割方法。实验表明，与传统的Ｋ平均方法相比，该方法除具有自动分割优点外，还具有稳定性好，自适应强，分割准确等优点。 Image segmentation is the most important and difficult problem in multidimensional ultrasound medical image reconstruction. In this paper, the traditional nearest neighbor classification method is combined with the self-organizing neural network to propose an automatic segmentation method of ultrasonic medical images. Experiments show that, compared with the traditional K-means, this method not only has the advantages of automatic segmentation, but also has the advantages of good stability, strong adaptability and accurate segmentation.