Predicate |
Object |
assignee |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_be2c55a9de09e08e3404c0a5a1ae4dc7 http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_a3e9438c2a1373794888bf3bd4f16807 |
classificationCPCAdditional |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-30204 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-30096 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-30092 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/G06T7-90 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-51 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-0012 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-11 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 |
classificationIPCInventive |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-11 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-00 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-90 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F16-51 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 |
filingDate |
2020-06-30-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
inventor |
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_571b36845e56d00abdc5577770b30e36 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_cb0bb9de0fc4d46160c7c0904896416a http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_b28fb699bb9cda3351990a80d8d248e2 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_bda9629add05315bb0e86b694d3bc9f7 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_783422408e9467f2dc774716fd261671 |
publicationDate |
2020-10-20-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber |
CN-111798427-A |
titleOfInvention |
A transfer learning-based mitotic detection system in gastrointestinal stromal tumors |
abstract |
A system for detecting mitoses in gastrointestinal stromal tumors based on transfer learning: preprocessing pathological slice image data; constructing a pretrained full convolutional neural network, including encoder sub-network and decoder sub-network; selecting Adam to optimize The pre-trained fully convolutional neural network is updated by the gradient controller, and the Focal Loss loss function is used to train the pre-trained fully convolutional neural network; 3) The decoder sub-network weights in the pre-trained fully trained fully convolutional neural network are calculated. After initialization, a new fully convolutional neural network is obtained, the Adam optimizer is selected to update the gradient of the fully convolutional neural network, and the Focal Loss loss function is used to train the fully convolutional neural network; A good fully convolutional neural network is tested to obtain a binary map of the mitotic region. The present invention can provide effective intermediate data for pathologists to detect mitotic images, and realize rapid judgment of patient's condition. |
isCitedBy |
http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113327233-A |
priorityDate |
2020-06-30-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type |
http://data.epo.org/linked-data/def/patent/Publication |