http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114359310-A
Outgoing Links
Predicate | Object |
---|---|
assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_d6a6f422b091ba12ea61d4adbf1b0e8e |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-11 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 |
filingDate | 2022-01-13-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_253ab6132460aa71bc8031dc9d529f95 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_64cb025ccfe87e44bb9dbcdfd65b25d4 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_8c68be7a0a30d8f787c5360958f879b8 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_15cee23416f210ed2667e7dad7dac62c |
publicationDate | 2022-04-15-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-114359310-A |
titleOfInvention | 3D ventricle nuclear magnetic resonance video segmentation optimization system based on deep learning |
abstract | The invention discloses a 3D ventricle nuclear magnetic resonance video segmentation optimization system based on deep learning, which obtains high-dimensional image characteristics in an MRI video band through a depth space-time deformable convolution fusion module TDAM; utilizing space-time information in high-dimensional image features obtained by the enhanced deformable convolution attention network TDAM to output a feature map with multi-scale information after fusing feature maps with different scales; obtaining the distribution of high-dimensional image characteristics through a probability noise correction module PNCM (public network communication module), and outputting an embedded vector containing distribution mean and variance information; and performing splicing convolution after the EDAN output characteristic diagram and the PNCM output embedded vector are expanded to obtain a prediction result. The newly input 3D ventricle MRI video is directly segmented by utilizing the trained network model, and the accuracy and efficiency of ventricle segmentation can be effectively improved by introducing multi-frame image compensation, deformable convolution and a multi-scale attention mechanism, and the method has higher robustness. |
isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-116152285-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-116152285-B |
priorityDate | 2022-01-13-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 21.