| Predicate |
Object |
| classificationCPCAdditional |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20076
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-10081
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-30016
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20084
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20081 |
| classificationCPCInventive |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-11
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06V10-44
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-241
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-0012
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-08
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-2415
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-047
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-048
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-33 |
| classificationIPCInventive |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-46
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-33
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-11
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-00 |
| filingDate |
2021-06-29-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| grantDate |
2021-09-07-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationDate |
2021-09-07-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber |
CN-113177943-B |
| titleOfInvention |
A CT image segmentation method for stroke |
| abstract |
The invention discloses a stroke CT image segmentation method, which includes inverting the ischemic stroke CT image, preprocessing the original CT image and the inverted CT image; constructing a twin multi-level encoder, and a calculation module to calculate two The feature difference of each level of the encoder is used to fuse the features with a multi-level feature fusion module; a shared decoder is constructed; a joint loss function is designed to train the optimal segmentation model on the training set; finally, the trained segmentation model is used to segment the unknown segmentation. Labeled test set for ischemic stroke infarct segmentation. The invention uses the feature difference calculation module to calculate the feature difference of each level of the two encoders, and uses the multi-level feature fusion module to fuse the global and local features; the infarction in the CT image can be more accurately segmented, and the ischemia is improved. It provides technical support and reference for the diagnosis efficiency and accuracy of cerebral apoplexy, reducing the fatality rate and disability rate. |
| priorityDate |
2021-06-29-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| type |
http://data.epo.org/linked-data/def/patent/Publication |