Open AccessSegmentation and 3D Reconstruction of Cerebral Hematoma Based on Deep Learning
Author(s) -
Taipeng Guo
Publication year - 2021
Publication title -
converter
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.104
H-Index - 1
ISSN - 0010-8189
DOI - 10.17762/converter.80
Subject(s) - hematoma , grasp , artificial intelligence , skull , segmentation , computer science , artificial neural network , medicine , epidural hematoma , cerebral trauma , deep learning , data set , set (abstract data type) , radiology , computer vision , surgery , programming language
Cerebral hemorrhage is a common clinical disease. Because of its rapid onset, high mortality and disability rate, in the treatment of cerebral hemorrhage, it is very important to accurately calculate the brain hematoma volume and feedback its location information in a short period of time. This paper proposes a method for precise segmentation and three-dimensional reconstruction of cerebral hematoma area based on deep learning. This method highlights the image information by expanding the CT image and eliminating the skull information, then accurately segments the cerebral hematoma areathrough the neural network model to build a three-dimensional model. We verify the experimental results based on the data set collected by the Affiliated Hospital of Xiangnan University, which proves the effectiveness of this method and its ability to significantly improve the speed incerebral hemorrhage area judgment and grasp information in clinical diagnosis.