首页|Northeast Electric Power University Reports Findings in Neural Computation (A Sp inal MRI Image Segmentation Method Based on Improved Swin-UNet)
Northeast Electric Power University Reports Findings in Neural Computation (A Sp inal MRI Image Segmentation Method Based on Improved Swin-UNet)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Computation - Neural C omputation is the subject of a report. According to news reporting from Jilin, P eople’s Republic of China, by NewsRx journalists, research stated, “As the numbe r of patients increases, physicians are dealing with more and more cases of dege nerative spine pathologies on a daily basis. To reduce the workload of healthcar e professionals, we propose a modified Swin-UNet network model.” The news correspondents obtained a quote from the research from Northeast Electr ic Power University, “Firstly, the Swin Transformer Blocks are improved using a residual post-normalization and scaling cosine attention mechanism, which makes the training process of the model more stable and improves the accuracy. Secondl y, we use the log-space continuous position biasing method instead of the bicubi c interpolation position biasing method. This method solves the problem of perfo rmance loss caused by the large difference between the resolution of the pretrai ning image and the resolution of the spine image. Finally, we introduce a segmen tation smooth module (SSM) at the decoder stage. The SSM effectively reduces red undancy, and enhances the segmentation edge processing to improve the model’s se gmentation accuracy. To validate the proposed method, we conducted experiments on a real dataset provided by hospitals. The average segmentation accuracy is no less than 95%.”
JilinPeople’s Republic of ChinaAsiaComputationNeural Computation
NETL
NSTL
