»ã±¨±êÌâ (Title)£ºMedical Image Quality Enhancement and Segmentation with Deep Learning£¨»ùÓÚÉî¶È½ø½¨µÄҽѧͼÏñÖÊÁ¿¼ÓÇ¿ºÍÔ׸

»ã±¨ÈË (Speaker)£ºÀîÖÇ ÌØƸ×êÑÐÔ±£¨»ª¶«Ê¦·¶´óѧ£©

»ã±¨¹¦·ò (Time)£º2023Äê11ÔÂ2ÈÕ(ÖÜËÄ) 13:30

»ã±¨µØÖ· (Place)£ºÐ£±¾²¿ F420

Ô¼ÇëÈË(Inviter)£ºÎÂÖÇæ¼

Ö÷°ì²¿ÃÅ£ºÀíѧԺÊýѧϵ

»ã±¨ÌáÒª£ºIn this talk, we explore the domain of medical image analysis, where the influential impact of deep learning architectures comes to the forefront in addressing the intricate challenges posed by medical imaging. This methodology places significant emphasis on the vital necessity of capturing information at various scales to yield superior outcomes. Whether it's through the utilization of a recurrent gate module and a multi-scale module for polyp segmentation, the incorporation of an adaptive selection aggregation module to enhance feature fusion, or the integration of a dual-channel neural network that underscores the interactions between denoising and super-resolution tasks, which collectively underscore the transformative potential of deep learning.