基于仿生模式识别的医学图像分割方法

提出了一种基于仿生模式识别的医学图像分割算法。该算法首先根据训练样本矢量确定Ψ³神经元的权值，并在此基础上构建多权值神经元网络；然后利用神经元网络完成样本在高维特征空间的最佳覆盖；最后根据覆盖结果进行识别、分割。实验结果表明，与传统医学图像分割方法相比，该算法具有更高的准确性和可靠性，更好的泛化能力。此外，该算法从“认识”的角度出发，可以有效融合先验知识，能快速准确地从医学图像中分割出感兴趣的区域，具有较高的智能性。

Medical image segmentation based on biomimetic pattern recognition

A medical image segmentation algorithm based on biomimetic pattern recognition is proposed.First,Ψ³-neurons’ weights are determined according to training samples and then Multi-Weights Neuron Networks are established.Second,the neuron networks are used to completely cover sample’s high dimensional feature space.Finally,medical images are recognized and segmented based on the results of the optimal coverage of the samples.The experimental results show that the proposed method has higher accuracy,reliability and better generalization than the traditional medical image segmentation methods.Besides,from the cognitive respect,this algorithm emphasizes on“cognition”,which can effectively integrate transcendental knowledge and obtain the desired object from medical images quickly and reliably,therefore it is highly intelligent.