报告人：Haimiao Zhang（Peking University）
时间：2018-10-26 12:00-13:30
地点：Room 1560, Sciences Building No. 1
12:00-12:30 lunch；12:30-13:30 Talk
Abstract: CT image reconstruction from incomplete data, such as sparse views and limited angle
reconstruction, is an important and challenging problem in medical imaging. In this talk, I will
present a deep convolutional neural network (CNN) architecture, called JSR-Net, that jointly
reconstructs CT images and their associated Radon domain projections. JSR-Net combines the
traditional model based approach with deep architecture design of deep learning. A hybrid loss
function is adopted to improve the JSR-Net performance, which is efficient to protect important
structures in the phantom. Numerical experiments demonstrate that JSR-Net outperforms some
latest model based reconstruction methods, as well as a recently proposed deep model.
欢迎各位同学积极参加，报名链接https://www.wjx.top/jq/29438199.aspx， 报名截止时间为10月25日
下午16:00，我们将为报名同学提供午餐。
