A new deep learning algorithm is proposed by introducing new concepts such as rotation invariance and split dropout. We discuss the problems in medical image analysis from such aspects as image segmentation, image registration and pattern recognition, establish a series of new mathematical models and design a new algorithm on the basis of geometrical analysis. By correlating the research results, we develop a set of applied, credible system to treat with medical image. The system succeeded in being widely used in 12 renowned hospitals such as Beijing 301 hospital in China and saving lives.

This presentation is part of Minisymposium “MS22 - Mapping problems in imaging, graphics and vision (3 parts)”
organized by: Ronald Lui (Chinese University of Hong Kong) , Ke Chen (University of Liverpool) .

Authors:

Dexing Kong (Zhejiang University)

Keywords: