http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114403845-A
Outgoing Links
| Predicate | Object |
|---|---|
| assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_1e6a86e1278a5c9d5152ed303324fd28 |
| classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/A61B2503-40 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/A61B2503-42 |
| classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/A61B5-055 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/A61B5-0042 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/A61B5-7267 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/A61B5-4845 |
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/A61B5-055 |
| filingDate | 2021-12-31-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_8b78c8fb35b58220fa0cddc9bd70b6ce http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_12d823b96a14cc790ace00e7d619c9ac |
| publicationDate | 2022-04-29-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-114403845-A |
| titleOfInvention | Dynamic addiction neural loop generation method and system based on weak supervision contrast learning |
| abstract | The invention relates to a method and a system for generating a dynamic addiction neural loop based on weak supervision contrast learning, belonging to the technical field of artificial intelligence; the generation method comprises the following steps: based on a convolutional neural network, reducing dimensions of multiple groups of voxels of fMRI to brain region node attributes, and generating multiple groups of dynamic brain connection graphs containing time sequences according to the brain region node attributes; extracting spatiotemporal features of brain connections in each dynamic brain connection map; inputting the space-time characteristics into an abnormal connection detection network, calculating abnormal probability of brain connection based on contrast learning, and acquiring the brain connection with the maximum abnormal probability at each moment; and generating a dynamic addictive neural loop according to the neuroscience priori knowledge and the brain connection with the maximum abnormal probability. The fMRI images are directly input into the model, and redundant and complex preprocessing calculation is omitted. Through contrast learning, the difference of brain connections of different samples is obtained, and an addiction neural loop mechanism is disclosed by combining a small amount of neuroscience priori knowledge, so that the method is easy to train and high in precision. |
| priorityDate | 2021-12-31-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 21.