array(44) { ["TalkID"]=> string(4) "4008" ["Date"]=> string(10) "2017-12-13" ["talk_English_Title"]=> string(91) "3D Fully Convolutional Networks for Segmentation, Detection and Tracking in Medical Imaging" ["talk_Chinese_Title"]=> string(0) "" ["End_Time"]=> string(8) "17:20:00" ["Headline"]=> string(1) "0" ["Chinese_First_Name"]=> NULL ["Chinese_Last_Name"]=> NULL ["English_First_Name"]=> NULL ["English_Last_Name"]=> NULL ["English_Middle_Name"]=> NULL ["PID"]=> NULL ["Abstract_Chinese"]=> string(0) "" ["Abstract_English"]=> string(949) "Automated segmentation, detection and tracking are important yet challenging problems for medical image analysis and computer-aided diagnosis. Recent advances in deep learning have made it feasible to produce dense voxel-wise predictions of volumetric images that can be utilized for segmentation, detection, and tracking tasks. For example, multi-class 3D fully convolutional networks (FCNs) trained on manually labeled CT scans of several abdominal structures are now achieving state-of-the-art segmentation results without the need for handcrafting features or training organ-specific models. Similar 3D FCNs can be utilized for detection tasks, e.g. finding and segmenting lymph nodes in CT images, without much need for re-designing the network architecture. Furthermore, 3D FCNs can be integrated with image tracking algorithms in order to robustly track and segment challenging anatomical structures, like the airways of the lung in chest CT." ["Abstract_PDF"]=> string(0) "" ["PlaceID"]=> string(3) "101" ["Room_Description"]=> string(4) "R202" ["Room_Description_E"]=> string(4) "R202" ["Chinese_Name"]=> string(15) "天文數學館" ["English_Name"]=> string(34) "Astronomy and Mathematics Building" ["IID_1"]=> NULL ["IID_2"]=> NULL ["Time_S"]=> NULL ["Time_E"]=> NULL ["status"]=> NULL ["Shorthand"]=> NULL ["Shorthand1"]=> NULL ["PCAID"]=> NULL ["Gender"]=> NULL ["I_Title_C_1"]=> NULL ["I_Title_E_1"]=> NULL ["I_Title_Alt_1"]=> NULL ["Branch_1"]=> NULL ["Country_1"]=> NULL ["I_Title_C_2"]=> NULL ["I_Title_E_2"]=> NULL ["I_Title_Alt_2"]=> NULL ["Branch_2"]=> NULL ["Country_2"]=> NULL ["OrganizerID"]=> NULL ["CategoryID"]=> string(3) "225" ["Category"]=> string(8) "Workshop" ["Start_Time"]=> string(8) "16:30:00" ["Web"]=> string(15) "20171213_1.pptx" } Center of Advanced Study in Theoretical Sciences(CASTS)


Workshop on Artificial Intelligence for Medical Image Analysis

3D Fully Convolutional Networks for Segmentation, Detection and Tracking in Medical Imaging

Prof. Holger Roth

2017 - 12 - 13 (Wed.)
16:30 - 17:20
R202, Astronomy and Mathematics Building

Automated segmentation, detection and tracking are important yet challenging problems for medical image analysis and computer-aided diagnosis. Recent advances in deep learning have made it feasible to produce dense voxel-wise predictions of volumetric images that can be utilized for segmentation, detection, and tracking tasks. For example, multi-class 3D fully convolutional networks (FCNs) trained on manually labeled CT scans of several abdominal structures are now achieving state-of-the-art segmentation results without the need for handcrafting features or training organ-specific models. Similar 3D FCNs can be utilized for detection tasks, e.g. finding and segmenting lymph nodes in CT images, without much need for re-designing the network architecture. Furthermore, 3D FCNs can be integrated with image tracking algorithms in order to robustly track and segment challenging anatomical structures, like the airways of the lung in chest CT.



  繁體
中文