Research On Improved Fuzzy Cellular Neural Networks And Apply It To Medical Images Processing

White blood cell detection is one of the most basic and key steps in the automatic recognition system of white blood cells in microscopic blood images. Its accuracy and stability greatly affect the operating speed and recognition accuracy of the whole system. The method of Fuzzy Cellular Neural Network is applied to segmentation of microscopic white blood cell images in [1]. But we find it is not better to keep on detecting the edge of white blood cells. Because of this shortage of FCNN, we present an improved fuzzy cellular neural network (IFCNN) in paper. Experimental results show that IFCNN much less influence from factors such as edge integrity than the previously-used algorithms. It can detect almost all white blood cells and each cell is nearly complete.General image segmentation techniques in medical image processing seem to be theoretically applicable in the segmentation of serial CT liver images. In this paper, we focus our study on how to efficiently segment serial CT liver images using the neural-network based technique. Advantages of the FCNN based segmentation approach are its inherent connection with mathematical morphology, which is also an important tool in image processing applications. This approach has been successfully employed in white blood cell detection [1,6]. However, when applied in the segmentation of serial CT liver images, the obtained results are unsatisfactory, due to serial CT liver image's specific characteristics In this paper, FCNN is improved to be the novel neural network---Advanced Fuzzy Cellular Neural Network AFCNN. Just like FCNN, AFCNN still keeps its convergent property and global stability. When applied to segment serial CT liver images, AFCNN has the distinctive advantage over FCNN: it can keep boundary integrity better and have better recall accuracies such that the segmented images can approximate original liver images better.