http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111476859-A
Outgoing Links
Predicate | Object |
---|---|
assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_d6a6f422b091ba12ea61d4adbf1b0e8e |
classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-10104 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/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T19-003 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T11-003 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T11-00 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T19-00 |
filingDate | 2020-04-13-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_c116271c315d91f119770a49cd4bf703 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_2964f1fc94eadff7d373259f475a9f5c |
publicationDate | 2020-07-31-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-111476859-A |
titleOfInvention | A dynamic dual tracer PET imaging method based on 3D Unet |
abstract | The invention discloses a dynamic double-tracking PET imaging method based on 3D Unet. According to the 3D format of the double-tracking dynamic PET data, a targeted 3D convolution kernel is selected, and a feature extraction and reconstruction process is performed in the stereoscopic receptive field. Starting from the dynamic image sequence, the three-dimensional concentration distribution maps of two different single tracer PETs were directly reconstructed. The method of the invention realizes the reconstruction of the dynamic PET concentration distribution image of the mixed tracer through the three-dimensional Unet. It adopts a specific three-dimensional convolution kernel, and can simultaneously extract the features including spatial information and time information on the concentration distribution map; The skip connection structure splices the original output features of the down-sampling block with the reconstructed features of the subsequent corresponding up-sampling blocks, and further preserves the key details in the image; finally, the network is trained by combining the true value of a single tracer as a label and an error function. , to achieve accurate spatiotemporal simultaneous reconstruction of images. |
isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113379863-B http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113379863-A |
priorityDate | 2020-04-13-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 28.