http://rdf.ncbi.nlm.nih.gov/pubchem/patent/RU-2758338-C1
Outgoing Links
Predicate | Object |
---|---|
assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_15d950f0fa29f0c3087331bbbc6f05fe |
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/A61B5-00 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/A61B5-00 |
filingDate | 2020-12-18-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
grantDate | 2021-10-28-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_e114172f6557e27dd6b29217d3d526b2 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_12c53d2fc5787abb0454adbf5f002e63 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_45a331060f0658804a6d08237201b2aa http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_2d14156f5e4f033671f822dedd7b5038 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_6a048ff02499846421e67452b09faa93 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_87e0d03c025010e7dbf09481916fe1cd http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_24e94ce3cef3cfd004876dec292349b5 |
publicationDate | 2021-10-28-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | RU-2758338-C1 |
titleOfInvention | Method for choosing personalized antianemic therapy for patients with chronic renal failure older than 15 years |
abstract | FIELD: medicine; diagnostics.SUBSTANCE: analysis of the efficiency of previous therapy is performed, for this purpose, the sex is determined, while fertility is additionally determined for the female sex. Next, the patient’s age and body mass index are determined, following signs are determined in blood: the presence of a hepatitis B virus agent, a hepatitis C virus agent, an HIV agent, the current level of red blood cells (million/mcl), a value of change in the current level of red blood cells compared to the level of red blood cells a month ago, the current level of white blood cells in blood (thousand/mcl), a value of change in the current level of white blood cells compared to the level of white blood cells a month ago, the current level of platelets in blood (thousand/mcl), the current hematocrit number (%), a value of change in the current hematocrit number compared to the hematocrit number a month ago, the current hemoglobin level in blood (g/l), a value of change in the current hemoglobin level compared to the hemoglobin level a month ago, a value of change in the current hemoglobin level compared to the hemoglobin level 2 months ago, the transferrin saturation coefficient (CST) with iron (%), a value of change in the current CST with iron compared to the CST with iron 3 months ago, the current ferritin level (mcg/l), a value of change in the current ferritin level in blood compared to the ferritin level 3 months ago, the current albumin level (g/l), the current phosphatase level (U/l), the current level of total protein (g/l), the current calcium level (mmol/l), the current phosphorus level (mmol/l), the current potassium level (mmol/l), the current sodium level (mmol/l), the current iron level (mmol/l), the current level of urea before dialysis (mmol/l), a value of change in the current urea level compared to the previous level, the current glucose level (mmol/l), the current creatinine level (mcmol/l), the current level of vitamin D (ng/ml), the current cholesterol level (mmol/l), the current level of parathyroid hormone (pg/ml), the current level of plasma bicarbonate (mmol/l), a value of a CT/V indicator. Data on doses of drugs prescribed in the previous therapy is collected: the prescribed dose of epoetin alpha or epoetin beta administered intravenously (U/month), the prescribed dose of epoetin alpha or epoetin beta administered subcutaneously (U/month), the prescribed dose of darbepoetin alpha (mcg/month), the prescribed dose of methoxypolyethylene glycol-epoetin beta (mcg/month), the prescribed dose of iron (III) sucrose complex hydroxide (mg). The duration of action of prescribed erythropoietins is determined as: no, short, long. Then, a vector of the above-mentioned patient signs is formed, the vector of signs is entered into an input of a mathematical model for evaluating the efficiency of a treatment regimen, consisting of 131 decision trees educated using XGBoost algorithm. After that, the vector of patient signs is assigned to an “effective treatment regimen” class or to an “ineffective treatment regimen” class. When assigning the vector of signs to the “ineffective treatment regimen” class, it is additionally fed to the input of the mathematical model for determining the type of inefficiency of the treatment regimen, consisting of 227 decision trees educated using the XGBoost algorithm, which assigns this vector of signs to a “redundant treatment regimen” class or to an “ineffective treatment regimen due to failure to achieve targets” class. When assigning the vector of signs to the “redundant treatment regimen” class or to the “ineffective treatment regimen due to failure to achieve targets” class, a search is carried out for the most suitable treatment regimen, for this purpose, a vector of patient signs containing a certain set of signs is formed. Scaling of signs is carried out, after which the vector of scaled patient signs and vectors of scaled signs of anti-anemic treatment regimens from the database of effective anti-anemic treatment regimens containing anti-anemic treatment regimens previously marked up by nephrologists are fed to the input of the mathematical model representing a soft cosine measure operation based on a similarity matrix of drugs. Then, for each pairwise combination of the vector of scaled patient signs and vectors of scaled signs of previously marked anti-anemic treatment regimens, values of soft cosine measures are found from the database of effective anti-anemic treatment regimens, after finding values of measures, 3 vectors of signs corresponding to effective anti-anemic treatment regimens for specific patients are selected from the database of effective anti-anemic treatment regimens with the highest values of measures exceeding 0.8. After that, from these three treatment regimens corresponding to the above-mentioned 3 vectors of signs, the one is chosen with the largest soft cosine measure.EFFECT: method allows one to increase the accuracy of the sele |
priorityDate | 2020-12-18-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 124.