@Article{JICS-13-083, author = {Zhihui Cao, Yuhang Qin and Yunjie Chen}, title = {A Multi-Scale Fully Convolutional Networks Model for Brain MRI Segmentation}, journal = {Journal of Information and Computing Science}, year = {2018}, volume = {13}, number = {2}, pages = {083--088}, abstract = {1School of math and statistics, Nanjing University of Information Science & Technology, Nanjing 210044, China (Received October 01 2017, accepted January 15 2018) Abstract 。 Accurate  segmentation  for  brain  magnetic  resonance  (MR)  images  is  of  great  significance  to quantitative analysis of brain image. However, traditional segmentation methods suffer from the problems existing in brain images such as noise,  weak edges and intensity inhomogeneity (also named as  bias field). Convolutional neural networks based methods have been used to segment images; however, it is still hard to find accurate results for  brain  MR  images.  In  order  to  obtain  accurate  segmentation  results,  a  multi-scale  fully  convolution  networks model  (MSFCN)  is  proposed  in  this  paper.  First,  we  use  padding  convolutions  in  conv-layer  to  preserve  the resolution of feature maps. So we can obtain segmentation results with the same resolution as inputs. Then, different sized filters are utilized in the same conv-layer, after that, the outputs of these filters are concatenated together and fed  to  the  next  layer,  which  makes  the  model  learn  features  from  different  scales.  Both  experimental  results  and statistic results show that the proposed model can obtain more accurate results. }, issn = {3080-180X}, doi = {https://doi.org/}, url = {http://global-sci.org/intro/article_detail/jics/22450.html} }