《2016基于能量驱动的分水岭算法在MRI海马图像分割中的应用》.pdf

20 6 Vol20, No6 2008 6 JOURNAL OF COM PUT ERAIDED DESIGN COMPUTER GRAPH ICS June, 2008 MRI * 鲁 向 芦 勤 罗述谦 ( 100069) ( luxiangzz@ sohu. com) , , . MRI . , , . , . , . MRI , . ; ; ; TP391.4 1 The Application of EnergyDriven Watershed Transform Algorithm to Hippocampus Segmentation in MRI * Lu Xiang Lu Qin Luo Shuqian ( College of Biomedical E ngineer ing, Cap ital Medical University , Beij ing 100069) Abstract Traditional w atershed transform algorithm is flexible, but some regional characteristics in the boundary are seldom used. So, oversegmentation is serious in segmentation results. A method is presented, which employs energydriven w atershed transform algorithm to segment hippocampus in human brain M RI. For this method, the w atershed transform model is used. Water floods to each pixel along the shortest path from seed points, then computes the cost of flooding across the pixel. The cost can be used as the energy of this pixel. Driven by energy minimizing, the class to w hich the points in the contour belong is modified to make the contour develop into the ultimate result. Some inner characteristics and boundary condition are used to limit oversegmentation. The segmentation results show this method can be applied to the segmentation of some complex structures, such as hippocampus. Key ords hippocampus; w atershed transform algorithm; energydriven; topographical distance , , . , . , . , [ 3] , . , [ 12] . , - - - - : 200 10 12; : 2008 03 1 : (