WO2022225000A1 - 甲状腺疾患の診断に関する参考情報を提供する情報処理装置 - Google Patents
甲状腺疾患の診断に関する参考情報を提供する情報処理装置 Download PDFInfo
- Publication number
- WO2022225000A1 WO2022225000A1 PCT/JP2022/018395 JP2022018395W WO2022225000A1 WO 2022225000 A1 WO2022225000 A1 WO 2022225000A1 JP 2022018395 W JP2022018395 W JP 2022018395W WO 2022225000 A1 WO2022225000 A1 WO 2022225000A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- information
- subject
- trained model
- disease
- crea
- Prior art date
Links
- 230000010365 information processing Effects 0.000 title claims abstract description 86
- 208000024799 Thyroid disease Diseases 0.000 title description 15
- 208000021510 thyroid gland disease Diseases 0.000 title description 15
- 238000003745 diagnosis Methods 0.000 title description 4
- DDRJAANPRJIHGJ-UHFFFAOYSA-N creatinine Chemical compound CN1CC(=O)NC1=N DDRJAANPRJIHGJ-UHFFFAOYSA-N 0.000 claims abstract description 216
- 229940109239 creatinine Drugs 0.000 claims abstract description 108
- 102000011923 Thyrotropin Human genes 0.000 claims abstract description 83
- 108010061174 Thyrotropin Proteins 0.000 claims abstract description 83
- HVYWMOMLDIMFJA-DPAQBDIFSA-N cholesterol Chemical compound C1C=C2C[C@@H](O)CC[C@]2(C)[C@@H]2[C@@H]1[C@@H]1CC[C@H]([C@H](C)CCCC(C)C)[C@@]1(C)CC2 HVYWMOMLDIMFJA-DPAQBDIFSA-N 0.000 claims abstract description 68
- 102000003914 Cholinesterases Human genes 0.000 claims abstract description 50
- 108090000322 Cholinesterases Proteins 0.000 claims abstract description 50
- 229940048961 cholinesterase Drugs 0.000 claims abstract description 50
- 102000004420 Creatine Kinase Human genes 0.000 claims abstract description 48
- 108010042126 Creatine kinase Proteins 0.000 claims abstract description 48
- 235000012000 cholesterol Nutrition 0.000 claims abstract description 34
- 238000009534 blood test Methods 0.000 claims abstract description 33
- 102000004169 proteins and genes Human genes 0.000 claims abstract description 32
- 108090000623 proteins and genes Proteins 0.000 claims abstract description 32
- 208000023328 Basedow disease Diseases 0.000 claims description 66
- 208000015023 Graves' disease Diseases 0.000 claims description 63
- 206010043778 thyroiditis Diseases 0.000 claims description 51
- 102000002260 Alkaline Phosphatase Human genes 0.000 claims description 47
- 108020004774 Alkaline Phosphatase Proteins 0.000 claims description 47
- 206010020850 Hyperthyroidism Diseases 0.000 claims description 32
- 208000005057 thyrotoxicosis Diseases 0.000 claims description 32
- 201000010099 disease Diseases 0.000 claims description 22
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 22
- 239000005495 thyroid hormone Substances 0.000 claims description 22
- 229940036555 thyroid hormone Drugs 0.000 claims description 22
- AUYYCJSJGJYCDS-LBPRGKRZSA-N Thyrolar Chemical class IC1=CC(C[C@H](N)C(O)=O)=CC(I)=C1OC1=CC=C(O)C(I)=C1 AUYYCJSJGJYCDS-LBPRGKRZSA-N 0.000 claims description 20
- 210000000265 leukocyte Anatomy 0.000 claims description 14
- 238000008050 Total Bilirubin Reagent Methods 0.000 claims description 8
- 210000003651 basophil Anatomy 0.000 claims description 8
- 210000000440 neutrophil Anatomy 0.000 claims description 7
- 101710107035 Gamma-glutamyltranspeptidase Proteins 0.000 claims description 6
- 101710173228 Glutathione hydrolase proenzyme Proteins 0.000 claims description 6
- 102000006640 gamma-Glutamyltransferase Human genes 0.000 claims description 6
- 238000012360 testing method Methods 0.000 description 71
- 238000012545 processing Methods 0.000 description 65
- 230000002489 hematologic effect Effects 0.000 description 32
- 238000010801 machine learning Methods 0.000 description 30
- 235000018102 proteins Nutrition 0.000 description 27
- 238000000034 method Methods 0.000 description 24
- 238000004891 communication Methods 0.000 description 17
- 230000008569 process Effects 0.000 description 14
- 230000036541 health Effects 0.000 description 9
- 230000015654 memory Effects 0.000 description 8
- 238000012549 training Methods 0.000 description 8
- XUIIKFGFIJCVMT-GFCCVEGCSA-N D-thyroxine Chemical compound IC1=CC(C[C@@H](N)C(O)=O)=CC(I)=C1OC1=CC(I)=C(O)C(I)=C1 XUIIKFGFIJCVMT-GFCCVEGCSA-N 0.000 description 6
- 230000009471 action Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 229940034208 thyroxine Drugs 0.000 description 6
- XUIIKFGFIJCVMT-UHFFFAOYSA-N thyroxine-binding globulin Natural products IC1=CC(CC([NH3+])C([O-])=O)=CC(I)=C1OC1=CC(I)=C(O)C(I)=C1 XUIIKFGFIJCVMT-UHFFFAOYSA-N 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000003672 processing method Methods 0.000 description 5
- 102100036475 Alanine aminotransferase 1 Human genes 0.000 description 4
- 108010082126 Alanine transaminase Proteins 0.000 description 4
- 210000004369 blood Anatomy 0.000 description 4
- 239000008280 blood Substances 0.000 description 4
- 208000003532 hypothyroidism Diseases 0.000 description 4
- 230000002989 hypothyroidism Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 208000024891 symptom Diseases 0.000 description 4
- 210000001685 thyroid gland Anatomy 0.000 description 4
- 102000001554 Hemoglobins Human genes 0.000 description 3
- 108010054147 Hemoglobins Proteins 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 108010003415 Aspartate Aminotransferases Proteins 0.000 description 2
- 102000004625 Aspartate Aminotransferases Human genes 0.000 description 2
- BPYKTIZUTYGOLE-IFADSCNNSA-N Bilirubin Chemical compound N1C(=O)C(C)=C(C=C)\C1=C\C1=C(C)C(CCC(O)=O)=C(CC2=C(C(C)=C(\C=C/3C(=C(C=C)C(=O)N\3)C)N2)CCC(O)=O)N1 BPYKTIZUTYGOLE-IFADSCNNSA-N 0.000 description 2
- 102000003855 L-lactate dehydrogenase Human genes 0.000 description 2
- 108700023483 L-lactate dehydrogenases Proteins 0.000 description 2
- 108010072866 Prostate-Specific Antigen Proteins 0.000 description 2
- 102100038358 Prostate-specific antigen Human genes 0.000 description 2
- LEHOTFFKMJEONL-UHFFFAOYSA-N Uric Acid Chemical compound N1C(=O)NC(=O)C2=C1NC(=O)N2 LEHOTFFKMJEONL-UHFFFAOYSA-N 0.000 description 2
- TVWHNULVHGKJHS-UHFFFAOYSA-N Uric acid Natural products N1C(=O)NC(=O)C2NC(=O)NC21 TVWHNULVHGKJHS-UHFFFAOYSA-N 0.000 description 2
- 238000004820 blood count Methods 0.000 description 2
- 230000003111 delayed effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 210000003743 erythrocyte Anatomy 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 210000001616 monocyte Anatomy 0.000 description 2
- 210000002966 serum Anatomy 0.000 description 2
- 230000001225 therapeutic effect Effects 0.000 description 2
- 231100000399 thyrotoxic Toxicity 0.000 description 2
- 230000001897 thyrotoxic effect Effects 0.000 description 2
- 210000001519 tissue Anatomy 0.000 description 2
- 229940116269 uric acid Drugs 0.000 description 2
- ZCYVEMRRCGMTRW-UHFFFAOYSA-N 7553-56-2 Chemical compound [I] ZCYVEMRRCGMTRW-UHFFFAOYSA-N 0.000 description 1
- 206010003658 Atrial Fibrillation Diseases 0.000 description 1
- 206010007559 Cardiac failure congestive Diseases 0.000 description 1
- 206010010774 Constipation Diseases 0.000 description 1
- 206010012735 Diarrhoea Diseases 0.000 description 1
- 206010017076 Fracture Diseases 0.000 description 1
- 206010019280 Heart failures Diseases 0.000 description 1
- 206010022998 Irritability Diseases 0.000 description 1
- 206010024264 Lethargy Diseases 0.000 description 1
- 206010060819 Myxoedema coma Diseases 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 108090000279 Peptidyltransferases Proteins 0.000 description 1
- 102000004160 Phosphoric Monoester Hydrolases Human genes 0.000 description 1
- 108090000608 Phosphoric Monoester Hydrolases Proteins 0.000 description 1
- 206010060862 Prostate cancer Diseases 0.000 description 1
- 208000000236 Prostatic Neoplasms Diseases 0.000 description 1
- 208000032140 Sleepiness Diseases 0.000 description 1
- 206010041349 Somnolence Diseases 0.000 description 1
- 206010042674 Swelling Diseases 0.000 description 1
- 102000003911 Thyrotropin Receptors Human genes 0.000 description 1
- 108090000253 Thyrotropin Receptors Proteins 0.000 description 1
- 206010044565 Tremor Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000037396 body weight Effects 0.000 description 1
- 230000000747 cardiac effect Effects 0.000 description 1
- 238000003748 differential diagnosis Methods 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 210000003979 eosinophil Anatomy 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 206010016256 fatigue Diseases 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 125000002642 gamma-glutamyl group Chemical group 0.000 description 1
- 230000036449 good health Effects 0.000 description 1
- 238000005534 hematocrit Methods 0.000 description 1
- 230000001771 impaired effect Effects 0.000 description 1
- 229910052740 iodine Inorganic materials 0.000 description 1
- 239000011630 iodine Substances 0.000 description 1
- XMBWDFGMSWQBCA-RNFDNDRNSA-M iodine-131(1-) Chemical compound [131I-] XMBWDFGMSWQBCA-RNFDNDRNSA-M 0.000 description 1
- 210000004698 lymphocyte Anatomy 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000004060 metabolic process Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000004793 poor memory Effects 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 230000002285 radioactive effect Effects 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 210000004243 sweat Anatomy 0.000 description 1
- 230000008961 swelling Effects 0.000 description 1
- 238000009601 thyroid function test Methods 0.000 description 1
- 230000004304 visual acuity Effects 0.000 description 1
- 230000004584 weight gain Effects 0.000 description 1
- 235000019786 weight gain Nutrition 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/53—Immunoassay; Biospecific binding assay; Materials therefor
- G01N33/564—Immunoassay; Biospecific binding assay; Materials therefor for pre-existing immune complex or autoimmune disease, i.e. systemic lupus erythematosus, rheumatoid arthritis, multiple sclerosis, rheumatoid factors or complement components C1-C9
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/68—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving proteins, peptides or amino acids
- G01N33/6803—General methods of protein analysis not limited to specific proteins or families of proteins
- G01N33/6827—Total protein determination, e.g. albumin in urine
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/70—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving creatine or creatinine
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/74—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving hormones or other non-cytokine intercellular protein regulatory factors such as growth factors, including receptors to hormones and growth factors
- G01N33/76—Human chorionic gonadotropin including luteinising hormone, follicle stimulating hormone, thyroid stimulating hormone or their receptors
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/74—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving hormones or other non-cytokine intercellular protein regulatory factors such as growth factors, including receptors to hormones and growth factors
- G01N33/78—Thyroid gland hormones, e.g. T3, T4, TBH, TBG or their receptors
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/92—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving lipids, e.g. cholesterol, lipoproteins, or their receptors
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/40—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for data related to laboratory analysis, e.g. patient specimen analysis
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/10—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- C—CHEMISTRY; METALLURGY
- C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
- C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
- C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
- C12Q1/34—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving hydrolase
- C12Q1/44—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving hydrolase involving esterase
- C12Q1/46—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving hydrolase involving esterase involving cholinesterase
-
- C—CHEMISTRY; METALLURGY
- C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
- C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
- C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
- C12Q1/48—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving transferase
- C12Q1/50—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving transferase involving creatine phosphokinase
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2333/00—Assays involving biological materials from specific organisms or of a specific nature
- G01N2333/435—Assays involving biological materials from specific organisms or of a specific nature from animals; from humans
- G01N2333/575—Hormones
- G01N2333/59—Follicle-stimulating hormone [FSH]; Chorionic gonadotropins, e.g. HCG; Luteinising hormone [LH]; Thyroid-stimulating hormone [TSH]
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2333/00—Assays involving biological materials from specific organisms or of a specific nature
- G01N2333/90—Enzymes; Proenzymes
- G01N2333/914—Hydrolases (3)
- G01N2333/916—Hydrolases (3) acting on ester bonds (3.1), e.g. phosphatases (3.1.3), phospholipases C or phospholipases D (3.1.4)
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2800/00—Detection or diagnosis of diseases
- G01N2800/04—Endocrine or metabolic disorders
- G01N2800/046—Thyroid disorders
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/72—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving blood pigments, e.g. haemoglobin, bilirubin or other porphyrins; involving occult blood
- G01N33/728—Bilirubin; including biliverdin
Definitions
- the present invention relates to an information processing device, a program, and an information processing method that provide reference information for diagnosing thyroid disease.
- Patent Document 1 discloses a device for assessing the risk of prostate cancer based on blood marker information including free prostate-specific antigen (fPSA) and total PSA (tPSA) levels.
- fPSA free prostate-specific antigen
- tPSA total PSA
- GD Basedow's disease
- PT painless thyroiditis
- Diagnosis of Graves' disease (GD) includes serum free thyroxine (T4), free triiodothyronine (T3), serum thyroid-stimulating hormone (TSH) suppression, presence of thyroid-stimulating antibodies and/or increased radioiodine uptake confirmed by laboratory findings.
- Painless thyroiditis (PT) may also be differentiated from Graves' disease by having less radioactive iodine uptake.
- thyrotoxicosis vary from person to person, and thyroid function tests are not performed in regular health checkups, so the diagnosis may be delayed.
- delayed treatment of Graves' disease increases the risk of atrial fibrillation, congestive heart failure, and fractures, and early treatment of Graves' disease avoids severe cardiac or musculoskeletal complications. is also beneficial.
- the present invention has been made in view of the above problems, and provides reference information for diagnosing thyroid disease based on biochemical parameters such as age and weight, and the results of hematological tests performed in ordinary medical examinations. It is an object of the present invention to provide an information processing apparatus, a program, and an information processing method that can
- the information processing apparatus from the information on the subject's blood test, total protein (TP), cholinesterase (ChE), total cholesterol (TC), creatinine (CREA), and creatine phosphokinase ( CPK) is input to the first trained model, and a first determination unit outputs information regarding whether or not the subject's thyroid stimulating hormone (TSH) is within a range requiring treatment.
- TSH thyroid stimulating hormone
- a range of thyroid-stimulating hormone (TSH) that requires treatment refers to, for example, a thyroid-stimulating hormone (TSH) of 10 ⁇ IU/ml or more, but the range that requires treatment is limited to this. not something.
- the information processing apparatus based on the age of the subject and information on the blood test of the subject, total protein (TP), creatinine (CREA), neutrophils (Neu), and total a second discriminating unit that inputs information about bilirubin (T-Bill) into the second trained model and outputs information about whether the subject's thyroid stimulating hormone (TSH) is within the normal value range; have.
- TP total protein
- CREA creatinine
- Neu neutrophils
- T-Bill total protein
- T-Bill total protein
- T-Bill thyroid stimulating hormone
- a primary determination of whether a subject's thyroid stimulating hormone (TSH) is within the normal range is based on the results of a hematological test performed during a routine physical examination and the subject's age. information can be obtained.
- the range of thyroid-stimulating hormone (TSH) is within the normal range means, for example, that the thyroid-stimulating hormone (TSH) is 4.5 ⁇ IU/ml or less, and the range of normal values is It is not limited.
- normal values for thyroid stimulating hormone (TSH) can be set at 5.0 ⁇ IU/ml or less.
- the information processing apparatus from the age of the subject and information on the blood test of the subject, creatinine (CREA), total cholesterol (TC), alkaline phosphatase (ALP), and total protein (TP) into a third trained model, and outputs information as to whether or not the subject has Graves' disease.
- CREA creatinine
- TC total cholesterol
- ALP alkaline phosphatase
- TP total protein
- the primary information can be used as reference information as to whether or not to perform further examinations to confirm Graves' disease, and as reference information as to the progress of Graves' disease in the subject.
- the information processing apparatus obtains information on thyroid hormone FT3, thyroid hormone FT4, FT3/FT4, alkaline phosphatase (ALP), and creatinine (CREA) from information on a subject's blood test, It has a fourth discriminating unit for inputting to a fourth trained model and outputting information related to discriminating whether the subject is suffering from Graves' disease or painless thyroiditis.
- the thyroid hormone test results based on the information obtained from the thyroid hormone test results and the hematological test performed in a normal health checkup, it is possible to determine whether the subject is suffering from Graves' disease or painless thyroiditis.
- the primary information can be used as reference information regarding whether or not to perform further tests to confirm Graves' disease or painless thyroiditis.
- the information processing apparatus from information on a blood test of a subject, alkaline phosphatase (ALP), creatinine (CREA), total protein (TP), ⁇ -glutamyltranspeptidase ( ⁇ GTP), leukocytes ( WBC) is input to a fifth trained model, and a fifth discriminating unit for outputting information on discrimination between Graves' disease and painless thyroiditis as a disease that the subject is suffering from. .
- ALP alkaline phosphatase
- CREA creatinine
- TP total protein
- ⁇ GTP ⁇ -glutamyltranspeptidase
- WBC leukocytes
- the disease that the subject is suffering from is Graves' disease or painless thyroid gland. It is possible to obtain primary information on the discrimination of flames. In addition, the primary information can be used as reference information regarding whether or not to perform further tests to confirm Graves' disease or painless thyroiditis.
- the information processing apparatus from the age of the subject and information on the blood test of the subject, total cholesterol (TC), cholinesterase (ChE), creatinine (CREA), and creatine phosphokinase (CPK) information is input to a sixth trained model, and a sixth determination unit outputs information regarding whether or not the subject is suffering from painless thyroiditis.
- TC total cholesterol
- ChE cholinesterase
- CREA creatinine
- CPK creatine phosphokinase
- the primary information can be used as reference information regarding whether or not to perform further tests to confirm painless thyroiditis, and as reference information regarding the progress of painless thyroiditis in subjects.
- the information processing apparatus obtains creatinine (CREA), total cholesterol (TC), cholinesterase (ChE), creatine phosphokinase (CPK), and basophils from information on a blood test of a subject.
- the primary information can be used as reference information regarding whether or not to conduct further examinations to determine the details of thyrotoxicosis, and as reference information regarding the progress of the subject's thyrotoxicosis.
- the present invention provides a program for causing an information processing device to input and output each piece of information to the trained model in the first to seventh embodiments, and an information processing device to input each piece of information to the trained model. Also provided is an information processing method for performing the input and output of the .
- an information processing device capable of providing reference information for diagnosing a thyroid disease based on biochemical parameters such as age and the results of hematological tests performed in a normal health checkup. and an information processing method.
- FIG. 6 is an example of a flowchart illustrating machine learning processing performed by a learning unit; 6 is an example of a flowchart illustrating processing performed by a determination unit; It is an example of a processing sequence of a diagnostic system composed of a server and a user terminal.
- this embodiment an embodiment of the present invention (hereinafter referred to as “this embodiment”) will be described in detail, but the present invention is not limited to this, and various modifications are possible without departing from the gist thereof. is.
- Hypothyroidism refers to a state in which thyroid hormone action is lower than necessary in each tissue. Symptoms of hypothyroidism generally include lethargy, fatigue, swelling, sensitivity to cold, weight gain, slow movement, poor memory, and constipation. There are many. Severe hypothyroidism can lead to somnolence and impaired consciousness, sometimes called myxedema coma.
- Thyrotoxicosis refers to a condition in which the function of thyroid hormone is excessive in each tissue. In thyrotoxicosis, the metabolism becomes active, and the person becomes sensitive to heat and sweats profusely. In addition, because energy is wasted, the amount of food may increase, but the person may become thin. In addition, it also acts on the nerves and causes symptoms in various parts of the body such as hand tremors, irritability, and diarrhea. This thyrotoxicosis includes the following Graves' disease and painless thyroiditis.
- Glos' disease refers to a disease in which thyroid hormone is overproduced. Specifically, it refers to a disease in which the thyroid gland is stimulated by TSH receptor antibodies, resulting in excessive production of thyroid hormone.
- Panal thyroiditis refers to a condition in which thyroid hormones that have been stored in the thyroid gland leak into the blood for some reason, resulting in a temporary increase in thyroid hormones. Specifically, thyroid hormone leaks into the blood when thyroid follicles are destroyed.
- each disease is generally a disease that is diagnosed according to the guidelines published by the medical society concerning the disease, a disease that is described in the column of efficacy and effect in the package insert of ethical drugs, or , or at least any of the diseases understood as terms commonly used in the pharmaceutical and medical industries.
- Information on whether or not or “information on discrimination” may be, for example, information that it is healthy or normal, or the probability of being healthy or normal is 100% or normalized It may be information represented by a level. Further, the information may be information indicating that the person is suffering from a predetermined disease, or information that expresses the possibility of being affected by the predetermined disease by 100 percent or a standardized level. good. Furthermore, these information may include both information indicating that the patient is normal or in good health and information indicating that the patient is suffering from a disease.
- a qualitative biochemical parameter such as a biochemical parameter that is higher (e.g., more diseased) or lower (e.g., more normal or healthy) than a reference value for normal or healthy It may also contain quantitative information. Furthermore, these pieces of information may include the meaning of recommending further inspections such as detailed inspections.
- the information processing device 100 which is a thyroid disease determination device, controls total protein (TP), cholinesterase (ChE), total cholesterol (TC), creatinine (CREA), and creatine phosphokinase (CPK).
- TP total protein
- ChE cholinesterase
- TC total cholesterol
- CREA creatinine
- CPK creatine phosphokinase
- a diagnostic system 1 is constructed to determine whether a subject's thyroid-stimulating hormone (TSH) is within the therapeutic range or not, based on the information about.
- TSH thyroid-stimulating hormone
- FIG. 1 shows a block diagram showing an information processing device 100 included in the diagnostic system of this embodiment.
- the information processing apparatus 100 typically includes one or more processors 110, a communication interface 120 for controlling wired or wireless communication, an input/output interface 130, a memory 140, a storage 150, and interconnecting these components. , which cooperate to implement the processes, functions, or methods described in this disclosure.
- the processor 110 executes processes, functions, or methods implemented by codes or instructions contained in programs stored in the memory 140 .
- Processor 110 includes, by way of example and not limitation, one or more central processing units (CPUs), GPUs (Graphics Processing Units).
- the communication interface 120 transmits and receives various data to and from other information processing devices via a network.
- the communication may be performed by wire or wirelessly, and any communication protocol may be used as long as mutual communication can be performed.
- the communication interface 120 is implemented as hardware such as a network adapter, various types of communication software, or a combination thereof.
- the input/output interface 130 includes an input device for inputting various operations to the information processing device 100 and an output device for outputting processing results processed by the information processing device 100 .
- the input/output interface 130 includes an information input device such as a keyboard, mouse, and touch panel, a blood test device, and an image output device such as a display.
- the information processing apparatus 100 may receive a predetermined input by connecting the external input/output interface 130 .
- the information processing device 100 may be externally attached with a blood test device or the like.
- the memory 140 temporarily stores programs loaded from the storage 150 and provides a work area for the processor 110 .
- the memory 140 also temporarily stores various data generated while the processor 110 is executing the program.
- Memory 140 may be, by way of example and not limitation, high speed random access memory such as DRAM, SRAM, DDR RAM, or other random access solid state memory, or any combination thereof.
- Storage 150 stores programs, various functional units, and various data.
- Storage 150 may be, by way of example and not limitation, non-volatile memory such as magnetic disk storage, optical disk storage, flash memory devices, or other non-volatile solid-state storage, or the like, and may be combined.
- Other examples of storage 150 may include one or more storage devices remotely located from CPU 110 .
- storage 150 stores programs, functions and data structures, or subsets thereof.
- the information processing apparatus 100 functions as a determination unit 155, a display unit 156, and a learning unit 157 by the processor 110 executing instructions included in a program stored in the storage 150. It is configured.
- the operating system 151 includes, for example, procedures for handling various basic system services and performing tasks with the hardware.
- the network communication unit 152 connects the information processing device 100 to another computer, for example, via the communication interface 120 and one or more communication networks such as the Internet, other wide area networks, local area networks, and metropolitan area networks. used to
- the subject data 153 includes, for example, personal information such as subject's gender, age, and blood test results.
- the subject data 153 may be an electronic medical chart stored by the storage 150 or an electronic medical chart stored by a remote server.
- the learning data 154 is a data set used to generate a trained model, and the data set may be divided into teacher data and test data.
- the learning data 154 includes, for example, information on thyroid disease and thyroid-stimulating hormone, and information on blood tests.
- the learning data 154 may be stored in the storage 150 or in a remote server.
- the determination unit 155 executes a process of acquiring information related to the hematological test of the subject from the communication interface 120 or the input/output interface 130 .
- each information acquired by the determination unit 155 can be input to the information processing apparatus 100 via the communication interface 120 by a doctor, a nurse, other medical personnel, or a medical institution. Further, each information acquired by the determination unit 155 may be received by the determination unit 155 from another terminal or server via the input/output interface 130 .
- the determination unit 155 can store each acquired information in the subject data 153 .
- the determination unit used in the first embodiment will be referred to as a first determination unit 155
- the determination units used in the second to sixth embodiments described later will be referred to as a second determination unit 155 to a sixth determination unit 155, respectively.
- the discrimination section 155 when there is no need to distinguish between them, they are simply referred to as the discrimination section 155 .
- the trained models used in the first to sixth embodiments are referred to as the first trained model to the second trained model, respectively, and simply referred to as the trained model when there is no need to distinguish between them. .
- the hematological test information acquired by the first determination unit 155 includes, for example, the subject's total protein (TP), cholinesterase (ChE), total cholesterol (TC), creatinine (CREA), and creatine phosphokinase (CPK). ) and may contain other information.
- TP total protein
- ChE cholinesterase
- TC total cholesterol
- CREA creatinine
- CPK creatine phosphokinase
- the first determination unit 155 inputs the information on the hematological test of the subject acquired as described above to the first trained model, and determines whether the thyroid stimulating hormone (TSH) of the subject is within the range requiring treatment. Executes processing for outputting information about whether or not
- the first trained model is not particularly limited, but may be, for example, a trained model generated by machine learning processing based on the learning data 154 .
- the display unit 157 executes processing for controlling display of the information output by the first determination unit 155 on the display device. For example, the display unit 157 displays information regarding whether or not the subject's thyroid-stimulating hormone (TSH) is within a range that requires treatment, the probability that the subject's thyroid-stimulating hormone (TSH) is 10 ⁇ IU/ml or more, or 10 ⁇ IU /ml can be displayed and controlled on the display device.
- TSH thyroid-stimulating hormone
- the learning unit 158 executes processing for generating a trained model by machine learning processing based on the learning data 154 .
- Machine learning processes include supervised learning and semi-supervised learning.
- Trained Model As an example, a first trained model was created using Sony's Prediction One. For the creation of this first trained model, 3343 subjects with thyroid-stimulating hormone (TSH) levels exceeding 10 ⁇ IU/ml and 23716 subjects with thyroid-stimulating hormone (TSH) levels of 10 ⁇ IU/ml or less were used for training. We used the data.
- TSH thyroid-stimulating hormone
- TP total protein
- ChE cholinesterase
- TC total cholesterol
- CREA creatinine
- CPK creatine phosphokinase
- the data for each example may include thyroid-stimulating hormone (TSH) levels, blood test results, and the subject's age, sex, and weight.
- Hematology results include red blood cell count (RBC), hemoglobin content (Hb), hematocrit (Ht), mean corpuscular volume (MCV), mean corpuscular hemoglobin content (MCH), mean corpuscular hemoglobin concentration (MCHC), platelets count (Plt), white blood cell count (WBC), neutrophils (Neu), lymphocytes (Lym), monocytes (Mo), eosinophils (Eo), basophils (Ba), total protein (TP), Total Bilirubin (T-Bil), Aspartate Aminotransferase (AST), Alanine Aminotransferase (ALT), Lactate Dehydrogenase (LDH), ⁇ -Glutamyltranspeptidase ( ⁇ GTP), Alkaline Phosphatase (ALP), Cholinesterase (ChE), Crea
- Trained model (Modification 1) As another example, training data from 2427 subjects with thyroid stimulating hormone (TSH) levels exceeding 15 ⁇ IU/ml and 24,632 subjects with thyroid stimulating hormone (TSH) levels of 15 ⁇ IU/ml or less were used. 1 A trained model was also created.
- this learning data into a ratio of 7:3, prepared a data set of teacher data and test data, and created the first trained model using the teacher data.
- TP total protein
- ChE cholinesterase
- TC total cholesterol
- CREA creatinine
- CPK creatine phosphokinase
- the first trained model obtained in this way has a specificity, accuracy rate, and negative predictive value of 90% or more. ) is high, and it can be seen that about half of the patients who are in the range requiring treatment can be detected.
- FIG. 2 is a flowchart illustrating machine learning processing in which the learning unit 158 of the information processing apparatus 100 creates a trained model using teacher data.
- step S201 the learning unit 158 collects actual treatment data as learning data.
- the learning unit 158 collects actual treatment data from the subject data 153 of the information processing device 100 or subject data stored in another information processing device or server connected via a network, and collects actual treatment data for learning. Store in data 154 . It can be training data and test data.
- step S202 the learning unit 158 creates teacher data from the data set stored in the learning data 154, and uses the teacher data to provide information on whether or not thyroid stimulating hormone (TSH) is within a range that requires treatment. Create a first trained model that outputs For example, the learning unit 158 uses teacher data to perform machine learning using information about the value of thyroid-stimulating hormone (TSH) as a correct label.
- TSH thyroid stimulating hormone
- FIG. 3 is a flowchart illustrating a process in which the determination unit 155 of the information processing apparatus 100 outputs information about visual acuity prognosis when treatment is performed using a learned model.
- the determination unit 155 acquires information on the hematological test result of the subject.
- the determination unit 155 may acquire information about the result of the hematology test from another information processing device or measurement device via the communication interface 120, or may obtain information about the results of the hematology test via the input/output interface 130.
- You may acquire the result of a blood test from information input devices, such as , a touch panel.
- the determination unit 155 is scheduled to determine the subject's disease name or symptoms from electronic medical record information stored in the storage of the information processing device or in a server accessible by the information processing device. may obtain information about the treatment of
- step S302 the determination unit 155 inputs information about the result of the hematology test of the subject into the first trained model, and according to the first trained model, the thyroid stimulating hormone (TSH) is within the range requiring treatment. Outputs information about whether or not
- step S303 the display unit 156 controls display of the output information on an image output device such as a display.
- the display control method is not particularly limited, but for example, the possibility that thyroid-stimulating hormone (TSH) is in the range requiring treatment and the possibility that it is not in the range requiring treatment may be displayed in percentages of 100.
- TSH thyroid-stimulating hormone
- the thyroid stimulating hormone (TSH) value of the subject It is possible to obtain primary information as to whether or not is in the therapeutic range.
- the information processing device 100 which is a thyroid disease determination device, determines total protein (TP), creatinine (CREA), A diagnostic system 1 is constructed to determine whether a subject's thyroid stimulating hormone (TSH) is within the normal range based on information on neutral (Neu) and total bilirubin (T-Bill).
- TSH thyroid stimulating hormone
- the configuration, operation processing, etc. of the information processing apparatus 100 according to the second embodiment are basically the same as the configuration, operation processing, etc. of the information processing apparatus of the diagnostic system according to the first embodiment.
- the difference between the second embodiment and the first embodiment is that the information to be input to the second trained model is the age of the subject and the blood test information of the subject, including total protein (TP), creatinine ( CREA), neutrophils (Neu), and total bilirubin (T-Bill). The point is that it is information about whether it is within the range or not.
- the hematological test information acquired by the second determination unit 155 includes, for example, the subject's total protein (TP), creatinine (CREA), neutrophils (Neu), and total bilirubin (T-Bill). ) and may contain other information.
- the biochemical parameters acquired by the second determination unit 155 include the subject's age and may include other information.
- the second determination unit 155 inputs the information on the subject's age and hematological test acquired as described above to the second trained model, and determines that the subject's thyroid stimulating hormone (TSH) is within the normal value range Execute processing to output information about whether or not there is.
- the second trained model is not particularly limited, but may be, for example, a trained model generated by machine learning processing based on the learning data 154 .
- the display unit 157 executes processing for controlling display of the information output by the second determination unit 155 on the display device. For example, the display unit 157 displays information about whether the subject's thyroid stimulating hormone (TSH) is within the normal value range, the probability that the subject's thyroid stimulating hormone (TSH) is 4.5 ⁇ IU/ml or less, Alternatively, the display can be controlled on the display device as the probability of exceeding 4.5 ⁇ IU/ml.
- TSH thyroid stimulating hormone
- TSH thyroid stimulating hormone
- the learning unit 158 executes processing for generating a trained model by machine learning processing based on the learning data 154 .
- Machine learning processes include supervised learning and semi-supervised learning.
- Trained Model As an example, a second trained model was created using Sony's Prediction One. To create this second trained model, 7850 subjects with thyroid stimulating hormone (TSH) levels greater than 4.5 ⁇ IU/ml and 7850 subjects with thyroid stimulating hormone (TSH) levels between The following learning data of 19209 subjects were used.
- TSH thyroid stimulating hormone
- this learning data into a ratio of 7:3, prepared a data set of teacher data and test data, and created a second trained model using the teacher data.
- a second trained model was obtained with predictors of total protein (TP), age, creatinine (CREA), neutrophils (Neu), and total bilirubin (T-Bill).
- step S202 the learning unit 158 creates teacher data from the data set stored in the learning data 154, and uses the teacher data to obtain information about whether thyroid stimulating hormone (TSH) is within the normal value range. Create a second trained model that outputs For example, the learning unit 158 uses teacher data to perform machine learning using information about the value of thyroid-stimulating hormone (TSH) as a correct label.
- TSH thyroid stimulating hormone
- step S302 the determination unit 155 inputs information about the subject's hematological test results to the trained model, and determines whether or not the thyroid stimulating hormone (TSH) is within the normal value range by the trained model. Output information.
- TSH thyroid stimulating hormone
- step S303 the display unit 156 controls display of the output information on an image output device such as a display.
- the display control method is not particularly limited, but for example, the possibility that the thyroid stimulating hormone (TSH) is within the normal value range and the possibility that it is not within the normal value range may be displayed in percentages of 100.
- TSH thyroid stimulating hormone
- the information processing device 100 which is a device for determining thyroid disease, determines the age of the subject and the blood test information of the subject to determine creatinine (CREA), total cholesterol (TC), alkaline Based on information on phosphatase (ALP) and total protein (TP), a diagnostic system 1 is constructed to determine whether or not a subject has Graves' disease.
- CREA creatinine
- TC total cholesterol
- ALP alkaline
- TP total protein
- the configuration, operation processing, etc. of the information processing apparatus 100 according to the third embodiment are basically the same as the configuration, operation processing, etc. of the information processing apparatus of the diagnostic system according to the first embodiment.
- the difference between the third embodiment and the first embodiment is that the information to be input to the third trained model is the age of the subject, and information on the subject's blood test such as creatinine (CREA), total cholesterol (TC), alkaline phosphatase (ALP), and total protein (TP), and the information output by the third trained model is information on whether the subject is suffering from Graves' disease. It is a point.
- the hematological test information acquired by the third determination unit 155 includes, for example, the subject's creatinine (CREA), total cholesterol (TC), alkaline phosphatase (ALP), and total protein (TP). and may contain other information.
- the biochemical parameters acquired by the third determination unit 155 include the subject's age and may include other information.
- the third discriminating unit 155 inputs the information about the subject's age and hematological test acquired as described above to the third trained model, and determines whether the subject is suffering from Graves' disease. Execute the output process.
- the third trained model is not particularly limited, but may be, for example, a trained model generated by machine learning processing based on the learning data 154 .
- the display unit 157 executes processing for controlling display of the information output by the third determination unit 155 on the display device.
- the display unit 157 can control and display information regarding whether or not the subject has Graves' disease on the display device as the probability of having Graves' disease or the probability of not having Graves' disease.
- the learning unit 158 executes processing for generating a trained model by machine learning processing based on the learning data 154 .
- Machine learning processes include supervised learning and semi-supervised learning.
- Trained Model As an example, a third trained model was created using Sony's Prediction One. Learning data of 19,335 untreated Graves' disease patients and 4,159 healthy subjects were used to create this third trained model.
- this learning data into a ratio of 7:3, prepared a data set of teacher data and test data, and created a third trained model using the teacher data.
- a third trained model was obtained with the subject's creatinine (CREA), total cholesterol (TC), alkaline phosphatase (ALP), age, and total protein (TP) as predictors.
- Trained model (Modification 1) Furthermore, in order to confirm the gender dependence of the determination of whether or not the subject is suffering from Graves' disease, the training data was divided into males and females, and training data and test data sets were prepared, A third trained model (female) and a third trained model (male) were created using teacher data.
- the third trained model female
- a model was obtained in which creatinine (CREA), total cholesterol (TC), cholinesterase (ChE), alkaline phosphatase (ALP), and creatine phosphokinase (CPK) were the predictors.
- CREA creatinine
- TC total cholesterol
- ChoE cholinesterase
- ALP alkaline phosphatase
- CPK creatine phosphokinase
- a model was obtained with alkaline phosphatase (ALP), creatinine (CREA), total cholesterol (TC), total protein (TP), and alanine aminotransferase (ALT) as predictors. was taken.
- the third trained model of the third embodiment may be replaced with the third trained model (female) or the third trained model (male).
- Trained model (Modification 2) Furthermore, in order to confirm the dependence on the free thyroxine (FT4) level in determining whether or not the subject is suffering from Graves' disease, from the above learning data, the free thyroxine (FT4) level was 5 ng / dl Test data for creating the above 3rd trained model (severe GD) for determining severe thyrotoxicosis and 3rd trained model for determining mild thyrotoxicosis with a free thyroxine (FT4) level of less than 5 ng/dl Created test data for creating a model (mild GD).
- test data for creating the third trained model were supervised data including 6040 severe thyrotoxicosis patients with free thyroxine (FT4) levels of 5 ng/dl or higher and 2911 healthy subjects.
- FT4 free thyroxine
- test data for creating the third trained model is training data including 7505 patients with mild thyrotoxicosis whose free thyroxine (FT4) level is less than 5 ng/dl and 2911 healthy subjects. , included test data from 3275 randomly selected thyrotoxic patients and 1248 healthy subjects.
- TC total cholesterol
- CREA creatinine
- TP total protein
- ChoE cholinesterase
- Mo monocyte
- a third trained model (mild GD) obtained a model with creatinine (CREA), total cholesterol (TC), age, alkaline phosphatase (ALP), and total protein (TP) as predictors.
- the ROC curve and AUC were obtained from the above test data.
- the third trained model of the third embodiment may be replaced with the third trained model (severe GD) or the third trained model (light GD).
- step S202 the learning unit 158 creates teacher data from the data set stored in the learning data 154, and uses the teacher data to output information regarding whether or not the subject has Graves' disease.
- 3 Create a trained model.
- the learning unit 158 uses teacher data to perform machine learning using information about whether or not the subject has Graves' disease as a correct label.
- step S302 the determination unit 155 inputs information about the subject's hematological test results to the third trained model, and determines whether the subject has Graves' disease based on the third trained model. Output information.
- step S303 the display unit 156 controls display of the output information on an image output device such as a display.
- the display control method is not particularly limited, for example, the possibility that the subject has Graves' disease and the possibility that the subject does not have Graves' disease may be indicated by 100%.
- the information processing apparatus 100 which is a thyroid disease determination apparatus, extracts thyroid hormone FT3, thyroid hormone FT4, FT3/FT4, alkaline phosphatase (ALP), and Information on creatinine (CREA) is input to the fourth trained model to construct a diagnostic system 1 that distinguishes whether the subject is suffering from Graves' disease or painless thyroiditis.
- the configuration, operation processing, etc. of the information processing apparatus 100 according to the fourth embodiment are basically the same as the configuration, operation processing, etc. of the information processing apparatus of the diagnostic system according to the first embodiment.
- the difference between the fourth embodiment and the first embodiment is that the information to be input to the fourth trained model includes thyroid hormone FT3, thyroid hormone FT4, FT3/FT4, alkaline phosphatase ( ALP) and creatinine (CREA), and the information output by the fourth trained model distinguishes whether the disease that the subject is suffering from is Basedow's disease or painless thyroiditis.
- the information to be input to the fourth trained model includes thyroid hormone FT3, thyroid hormone FT4, FT3/FT4, alkaline phosphatase ( ALP) and creatinine (CREA), and the information output by the fourth trained model distinguishes whether the disease that the subject is suffering from is Basedow's disease or painless thyroiditis.
- ALP alkaline phosphatase
- CREA creatinine
- the hematological test information acquired by the fourth determination unit 155 includes, for example, the subject's thyroid hormone FT3, thyroid hormone FT4, FT3/FT4, alkaline phosphatase (ALP), and creatinine (CREA). and may contain other information.
- the fourth determination unit 155 inputs the information on the subject's age and hematological test acquired as described above to the fourth trained model, and determines whether the subject's disease is Graves' disease or whether it is painless. Execute processing for outputting information related to discrimination of thyroiditis.
- the fourth trained model is not particularly limited, but may be, for example, a trained model generated by machine learning processing based on the learning data 154 .
- the display unit 157 executes processing for controlling display of the information output by the fourth determination unit 155 on the display device.
- the display unit 157 displays information related to distinguishing whether the subject is suffering from Graves' disease or painless thyroiditis.
- the display can be controlled on the display device as the probability of suffering from painless thyroiditis or the probability of not suffering from painless thyroiditis.
- the learning unit 158 executes processing for generating a trained model by machine learning processing based on the learning data 154 .
- Machine learning processes include supervised learning and semi-supervised learning.
- Trained Model As an example, a fourth trained model was created using Sony's Prediction One. Learning data of 19,335 untreated Graves' disease patients and 3,267 untreated painless thyroiditis patients were used to create this fourth trained model.
- this learning data into a ratio of 7:3, prepared a data set of teacher data and test data, and created a fourth trained model using the teacher data.
- a fourth trained model was obtained with thyroid hormone FT3, thyroid hormone FT4, FT3/FT4, alkaline phosphatase (ALP), and creatinine (CREA) as predictors.
- step S202 the learning unit 158 creates teacher data from the data set stored in the learning data 154, and uses the teacher data to determine whether the subject is suffering from Graves' disease or painless thyroiditis. Create a fourth trained model that outputs information on whether or not to discriminate.
- the learning unit 158 uses teacher data to perform machine learning using information about a disease (Graves' disease or painless thyroiditis) that the subject is suffering from as a correct label.
- step S302 the determination unit 155 inputs information about the result of the hematology test of the subject to the fourth trained model, and determines that Graves' disease is the disease that the subject is suffering from according to the fourth trained model. or painless thyroiditis.
- step S303 the display unit 156 controls display of the output information on an image output device such as a display.
- the display control method is not particularly limited. , and the probability of not being affected may be expressed as a percentage of 100.
- the information processing apparatus 100 which is a thyroid disease determination apparatus, extracts alkaline phosphatase (ALP), creatinine (CREA), total protein (TP), ⁇ -glutamyl from information on a blood test of a subject.
- a diagnostic system for inputting information on transpeptidase ( ⁇ GTP) and white blood cells (WBC) into a fifth trained model to distinguish whether the subject is suffering from Graves' disease or painless thyroiditis. 1 is constructed.
- the configuration, operation processing, etc. of the information processing apparatus 100 according to the fifth embodiment are basically the same as the configuration, operation processing, etc. of the information processing apparatus of the diagnostic system according to the first embodiment.
- the difference between the fifth embodiment and the first embodiment is that the information to be input to the fifth trained model is alkaline phosphatase (ALP), creatinine (CREA), total protein (TP), ⁇ -glutamyltranspeptidase ( ⁇ GTP), and white blood cells (WBC), and the information output by the fifth trained model is whether the subject is suffering from Graves' disease or is indolent.
- ALP alkaline phosphatase
- CREA creatinine
- TP total protein
- ⁇ GTP ⁇ -glutamyltranspeptidase
- WBC white blood cells
- the information on the hematological test acquired by the fifth determination unit 155 includes, for example, subject's alkaline phosphatase (ALP), creatinine (CREA), total protein (TP), ⁇ -glutamyltranspeptidase ( ⁇ GTP), White blood cells (WBC) are included and other information may be included.
- ALP alkaline phosphatase
- CREA creatinine
- TP total protein
- ⁇ GTP ⁇ -glutamyltranspeptidase
- WBC White blood cells
- the fifth determination unit 155 inputs the information on the subject's age and hematological test acquired as described above to the fifth trained model, and determines whether the subject's disease is Graves' disease or whether it is painless. Execute processing for outputting information related to discrimination of thyroiditis.
- the fifth trained model is not particularly limited, but may be, for example, a trained model generated by machine learning processing based on the learning data 154 .
- the display unit 157 executes processing for controlling display of the information output by the fifth determination unit 155 on the display device.
- the display unit 157 displays information related to distinguishing whether the subject is suffering from Graves' disease or painless thyroiditis.
- the display can be controlled on the display device as the probability of suffering from painless thyroiditis or the probability of not suffering from painless thyroiditis.
- the learning unit 158 executes processing for generating a trained model by machine learning processing based on the learning data 154 .
- Machine learning processes include supervised learning and semi-supervised learning.
- Trained Model As an example, a fifth trained model was created using Sony's Prediction One. Learning data of 19,335 untreated Graves' disease patients and 3,267 untreated painless thyroiditis patients were used to create this fifth trained model.
- this learning data into a ratio of 7:3, prepared a data set of teacher data and test data, and created a fifth trained model using the teacher data.
- a fifth trained model with alkaline phosphatase (ALP), creatinine (CREA), total protein (TP), ⁇ -glutamyltranspeptidase ( ⁇ GTP), and white blood cells (WBC) as predictors was obtained.
- ALP alkaline phosphatase
- CREA creatinine
- TP total protein
- ⁇ GTP ⁇ -glutamyltranspeptidase
- WBC white blood cells
- step S202 the learning unit 158 creates teacher data from the data set stored in the learning data 154, and uses the teacher data to determine whether the subject is suffering from Graves' disease or painless thyroiditis. Create a fifth trained model that outputs information on whether or not to discriminate.
- the learning unit 158 uses teacher data to perform machine learning using information about a disease (Graves' disease or painless thyroiditis) that the subject is suffering from as a correct label.
- step S302 the determination unit 155 inputs information about the result of the hematology test of the subject to the fifth trained model, and determines that Graves' disease is the disease that the subject is suffering from according to the fifth trained model. or painless thyroiditis.
- step S303 the display unit 156 controls display of the output information on an image output device such as a display.
- the display control method is not particularly limited. , the probability of not being affected may be expressed as a percentage of 100.
- the information processing apparatus 100 which is a thyroid disease determination apparatus, determines total cholesterol (TC), cholinesterase (ChE), and creatinine from the subject's age and blood test information. (CREA) and creatine phosphokinase (CPK) are input to the sixth trained model to construct a diagnostic system 1 that determines whether or not the subject is suffering from painless thyroiditis.
- TC total cholesterol
- ChE cholinesterase
- CPK creatinine phosphokinase
- the configuration, operation processing, etc. of the information processing apparatus 100 according to the sixth embodiment are basically the same as the configuration, operation processing, etc. of the information processing apparatus of the diagnostic system according to the first embodiment.
- the difference between the sixth embodiment and the first embodiment is that the information to be input to the sixth trained model is the age of the subject and the blood test information of the subject, including total cholesterol (TC), cholinesterase ( ChE), creatinine (CREA), and creatine phosphokinase (CPK), and the information output by the sixth trained model is information on whether the subject is suffering from painless thyroiditis This is the point.
- the hematological test information acquired by the sixth determination unit 155 includes, for example, the subject's total cholesterol (TC), cholinesterase (ChE), creatinine (CREA), and creatine phosphokinase (CPK). and may contain other information.
- the biochemical parameters acquired by the sixth determination unit 155 include the subject's age and may include other information.
- the sixth determination unit 155 inputs the information on the subject's age and hematological test acquired as described above to the sixth trained model, and determines whether the subject is suffering from painless thyroiditis. Execute a process that outputs information.
- the sixth trained model is not particularly limited, but may be, for example, a trained model generated by machine learning processing based on the learning data 154 .
- the display unit 157 executes processing for controlling display of the information output by the sixth determination unit 155 on the display device.
- the display unit 157 can display and control information on whether or not the subject has painless thyroiditis on the display device as the probability of having painless thyroiditis or the probability of not having painless thyroiditis. can.
- the learning unit 158 executes processing for generating a trained model by machine learning processing based on the learning data 154 .
- Machine learning processes include supervised learning and semi-supervised learning.
- 6.2. Trained Model As an example, a sixth trained model was created using Sony's Prediction One. The training data of 3267 untreated patients with painless thyroiditis and 4159 healthy subjects were used to create this sixth trained model.
- this learning data into a ratio of 7:3, prepared a data set of teacher data and test data, and created a sixth trained model using the teacher data.
- a sixth trained model was obtained with total cholesterol (TC), cholinesterase (ChE), creatinine (CREA), age, and creatine phosphokinase (CPK) as predictors.
- step S202 the learning unit 158 creates teacher data from the data set stored in the learning data 154, and uses the teacher data to output information regarding whether or not the subject is suffering from painless thyroiditis.
- Create a sixth trained model that For example, the learning unit 158 uses teacher data to perform machine learning using information about whether the subject is suffering from painless thyroiditis as a correct label.
- step S302 the determination unit 155 inputs information about the subject's hematological test results to the sixth trained model, and determines whether or not the subject is suffering from painless thyroiditis by the sixth trained model. Prints information about whether
- step S303 the display unit 156 controls display of the output information on an image output device such as a display.
- the display control method is not particularly limited, for example, the possibility that the subject is suffering from painless thyroiditis and the possibility that the subject is not suffering from painless thyroiditis may be displayed in percentages of 100.
- the information processing device 100 which is a thyroid disease determination device, extracts creatinine (CREA), total cholesterol (TC), cholinesterase (ChE), creatine phosphokinase from information on a blood test of a subject. (CPK) and information on basophils (Ba) are input to the seventh trained model to construct a diagnostic system 1 that determines whether or not the subject is suffering from thyrotoxicosis.
- the configuration, operation processing, etc. of the information processing apparatus 100 according to the seventh embodiment are basically the same as the configuration, operation processing, etc. of the information processing apparatus of the diagnostic system according to the first embodiment.
- the difference between the seventh embodiment and the first embodiment is that the information to be input to the seventh trained model is creatinine (CREA), total cholesterol (TC), cholinesterase (ChE) as information on the blood test of the subject. , creatine phosphokinase (CPK), and basophil (Ba), and the information output by the 7th trained model is information on whether the subject suffers from thyrotoxicosis.
- CREA creatinine
- TC total cholesterol
- Cholinesterase Cholinesterase
- CPK creatine phosphokinase
- Ba basophil
- the same reference numerals are given to the functional configurations, actions, and effects that are the same as those of the first embodiment, and the description is omitted, and the differences between the seventh embodiment and the first embodiment are mainly described. to explain.
- the hematological test information acquired by the seventh determination unit 155 includes, for example, the subject's creatinine (CREA), total cholesterol (TC), cholinesterase (ChE), creatine phosphokinase (CPK), A base ball (Ba) is included and other information may be included.
- CREA creatinine
- TC total cholesterol
- ChE cholinesterase
- CPK creatine phosphokinase
- Ba A base ball is included and other information may be included.
- the seventh determination unit 155 inputs the information about the subject's age and hematological test acquired as described above into the seventh trained model, and determines whether the subject suffers from thyrotoxicosis. Execute the process that outputs the .
- the seventh trained model is not particularly limited, but may be, for example, a trained model generated by machine learning processing based on the learning data 154 .
- the display unit 157 executes processing for controlling display of the information output by the seventh determination unit 155 on the display device.
- the display unit 157 can display and control information regarding whether or not the subject has thyrotoxicosis on the display device as the probability of having thyrotoxicosis or the probability of not having thyrotoxicosis.
- the learning unit 158 executes processing for generating a trained model by machine learning processing based on the learning data 154 .
- Machine learning processes include supervised learning and semi-supervised learning.
- Trained Model As an example, a seventh trained model was created using Sony's Prediction One. The training data of 19,335 untreated Graves' disease patients, 3,267 untreated painless thyroiditis patients, and 4,159 healthy subjects were used to create this seventh trained model.
- this learning data into a ratio of 7:3, prepared a data set of teacher data and test data, and created a seventh trained model using the teacher data.
- a seventh trained model was obtained with creatinine (CREA), total cholesterol (TC), cholinesterase (ChE), creatine phosphokinase (CPK), and basophils (Ba) as predictors.
- step S202 the learning unit 158 creates teacher data from the data set stored in the learning data 154, and uses the teacher data to output information about whether the subject is suffering from thyrotoxicosis. Create a seventh trained model. For example, the learning unit 158 uses teacher data to perform machine learning using information about whether the subject is suffering from thyrotoxicosis as a correct label.
- step S302 the determination unit 155 inputs information about the result of the hematological test of the subject to the seventh trained model, and determines whether the subject suffers from thyrotoxicosis by the seventh trained model. Prints information about Further, instead of or in addition to this, the determination unit 155 may output information regarding whether or not the thyroid stimulating hormone (TSH) is below the normal value by the seventh learned model.
- TSH thyroid stimulating hormone
- step S303 the display unit 156 controls display of the output information on an image output device such as a display.
- the display control method is not particularly limited, for example, the possibility that the subject is suffering from thyrotoxicosis and the possibility that the subject is not suffering from it may be displayed in 100% ratio.
- the diagnosis system 1 using the thyroid disease determination apparatus described in the first to seventh embodiments may be used independently or in combination.
- a combination of the first embodiment and the second embodiment, a combination of the third embodiment and the seventh embodiment, and a combination of the fourth embodiment and the seventh embodiment can be mentioned.
- First embodiment 1 1st Example 1 is an example which combined 1st Embodiment and 2nd Embodiment.
- the information processing apparatus 100 of the first embodiment calculates total protein (TP), creatinine (CREA), neutrophils (Neu), and total bilirubin (T- a second determination unit that inputs information about Bill) into a second trained model and outputs information about whether the subject's thyroid stimulating hormone (TSH) is within the normal value range; Input information on total protein (TP), cholinesterase (ChE), total cholesterol (TC), creatinine (CREA), and creatine phosphokinase (CPK) from blood test information into the first trained model, and a first determination unit that outputs information regarding whether the subject's thyroid stimulating hormone (TSH) is within a range requiring treatment.
- the configuration, operation processing, etc. of the information processing apparatus 100 according to the first embodiment are basically the same as the configuration, operation processing, etc. of the information processing apparatus of the diagnostic system according to the first embodiment and the second embodiment.
- the first embodiment 1 by combining the first embodiment and the second embodiment, it is possible to determine whether the subject's thyroid-stimulating hormone (TSH) is in the range requiring treatment or in the normal range. can be evaluated step by step.
- TSH thyroid-stimulating hormone
- the first example 2 is an example in which the third embodiment and the seventh embodiment are combined.
- the information processing apparatus 100 of the second embodiment obtains creatinine (CREA), total cholesterol (TC), cholinesterase (ChE), creatine phosphokinase (CPK), and basophils (Ba) from the information on the subject's blood test.
- CREA creatinine
- TC total cholesterol
- ChE cholinesterase
- CPK creatine phosphokinase
- Ba basophils
- a seventh determination unit that inputs information about whether the subject is suffering from thyrotoxicosis into the seventh learned model, outputs information about whether the subject is suffering from thyrotoxicosis, the subject's age, and the subject's blood Input information on creatinine (CREA), total cholesterol (TC), alkaline phosphatase (ALP), and total protein (TP) from information on tests into the third trained model, and the subject is diagnosed with Graves' disease and a third determination unit that outputs information regarding whether or not the determination is made.
- CREA creatinine
- TC total cholesterol
- ALP alkaline phosphatase
- TP total protein
- the configuration, operation processing, etc. of the information processing device 100 according to the first embodiment 2 are basically the same as the configuration, operation processing, etc. of the information processing device of the diagnostic system according to the third embodiment and the seventh embodiment.
- the first example 3 is an example in which the fourth embodiment and the seventh embodiment are combined.
- the information processing apparatus 100 of the third embodiment obtains creatinine (CREA), total cholesterol (TC), cholinesterase (ChE), creatine phosphokinase (CPK), and basophils (Ba) from the information on the subject's blood test.
- CREA creatinine
- TC total cholesterol
- ChE cholinesterase
- CPK creatine phosphokinase
- Ba basophils
- a seventh determination unit that inputs information about whether the subject is suffering from thyrotoxicosis into the seventh trained model and outputs information about whether the subject is suffering from thyrotoxicosis; , thyroid hormones FT4, FT3/FT4, alkaline phosphatase (ALP), and creatinine (CREA) are input into a fourth trained model, and whether the subject is suffering from Graves' disease or indolent and a fourth discriminating unit that outputs information related to discrimination of thyroiditis.
- the configuration, operation processing, etc. of the information processing device 100 according to the first embodiment 3 are basically the same as the configuration, operation processing, etc. of the information processing device of the diagnostic system according to the fourth embodiment and the seventh embodiment.
- the diagnostic system 1 using the thyroid disease determination device according to the first to seventh embodiments is a server (information 100) and the user terminal 200.
- FIG. 4 shows a processing sequence of a diagnostic system composed of a server (information processing device 100), which is a thyroid disease determination device, and a user terminal 200.
- the server 100 is communicably connected to the user terminal 200 via a communication network such as the Internet.
- the determination unit 155 of the server 100 inputs biochemical parameters such as the results of hematological tests and age received from the user terminal 200 into the learned model, A diagnostic system 1 that outputs each piece of information and transmits the output information to a user terminal 200 is provided.
- the server 100 is an example of an information processing device that implements all or part of the diagnostic system of the present invention, and can have the hardware configuration and functional unit configuration of the information processing device 100 described above.
- the user terminal 200 may be a normal computer equipped with a display capable of transmitting/receiving information and displaying information.
- step S401 the user terminal 200 transmits to the server 100 the results of hematological tests and information on biochemical parameters such as age. Obtain information on biochemical parameters such as age.
- step S402 the determination unit 155 of the server 100 inputs the information received from the user terminal 200 into the learned model, and outputs each result described in the first to seventh embodiments.
- step S403 the server 100 transmits the output result to the user terminal 200, and in step S404, the user terminal 200 controls the display of the received result on the display device.
- any user terminal 200 that can access the server 100 can use the diagnostic system of the present invention, thereby eliminating regional disparities in medical care levels.
- the present invention provides a program for causing an information processing device to input and output each piece of information to the trained model in the first to seventh embodiments, and an information processing device to input each piece of information to the trained model. Also provided is an information processing method for performing the input and output of the .
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Immunology (AREA)
- Urology & Nephrology (AREA)
- Hematology (AREA)
- Chemical & Material Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Physics & Mathematics (AREA)
- Medicinal Chemistry (AREA)
- Cell Biology (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Microbiology (AREA)
- General Physics & Mathematics (AREA)
- Biotechnology (AREA)
- Food Science & Technology (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Endocrinology (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Biophysics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Proteomics, Peptides & Aminoacids (AREA)
- Bioinformatics & Computational Biology (AREA)
- Rehabilitation Therapy (AREA)
- Rheumatology (AREA)
- Reproductive Health (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
Description
1.1.ハードウェア構成
第1実施形態では、甲状腺疾患判定装置である情報処理装置100によって、総プロテイン(TP)、コリンエステラーゼ(ChE)、総コレステロール(TC)、クレアチニン(CREA)、及びクレアチンホスホキナーゼ(CPK)に関する情報に基づいて、対象者の甲状腺刺激ホルモン(TSH)が治療を要する範囲であるか否かを判定する診断システム1が構築される。
一例として、Sony社のPrediction Oneを用いて第1学習済モデルを作成した。この第1学習済モデルの作成には、甲状腺刺激ホルモン(TSH)の値が10μIU/ml超過の被験者3343例と、甲状腺刺激ホルモン(TSH)の値が10μIU/ml以下の被験者23716例の学習用データを用いた。
AUC 0.737
さらに他の例として、甲状腺刺激ホルモン(TSH)の値が15μIU/ml超過の被験者2427例と、甲状腺刺激ホルモン(TSH)の値が15μIU/ml以下の被験者24632例の学習用データを用いた第1学習済モデルも作成した。
AUC 0.773
次に、このように構成された第1実施形態の情報処理装置100の動作について説明する。
図2は、情報処理装置100の学習部158が、教師データを用いて学習済モデルを作成する機械学習処理を例示するフローチャートである。
図3は、情報処理装置100の判別部155が、学習済モデルによって、治療を行った場合の視力予後に関する情報を出力する処理を例示するフローチャートである。
第2実施形態では、甲状腺疾患判定装置である情報処理装置100によって、対象者の年齢と、該対象者の血液検査に関する情報から、総プロテイン(TP)、クレアチニン(CREA)、好中球(Neu)、及び総ビリルビン(T-Bill)に関する情報に基づいて、対象者の甲状腺刺激ホルモン(TSH)が正常値の範囲であるか否かを判定する診断システム1が構築される。
第2判別部155が取得する血液学的検査に関する情報には、例えば、対象者の総プロテイン(TP)、クレアチニン(CREA)、好中球(Neu)、及び総ビリルビン(T-Bill)が含まれ、このほかの情報が含まれていてもよい。また、第2判別部155が取得する生化学的パラメータには、対象者の年齢が含まれ、このほかの情報が含まれていてもよい。
一例として、Sony社のPrediction Oneを用いて第2学習済モデルを作成した。この第2学習済モデルの作成には、甲状腺刺激ホルモン(TSH)の値が4.5μIU/ml超過の被験者7850例と、甲状腺刺激ホルモン(TSH)の値が0.2~4.5μIU/ml以下の被験者19209例の学習用データを用いた。
AUC 0.639
次に、このように構成された第2実施形態の情報処理装置100の動作について説明する。
第3実施形態では、甲状腺疾患判定装置である情報処理装置100によって、対象者の年齢と、該対象者の血液検査に関する情報から、クレアチニン(CREA)、総コレステロール(TC)、アルカリホスファターゼ(ALP)、及び総プロテイン(TP)に関する情報に基づいて、対象者がバセドウ病に罹患している否かを判定する診断システム1が構築される。
第3判別部155が取得する血液学的検査に関する情報には、例えば、対象者のクレアチニン(CREA)、総コレステロール(TC)、アルカリホスファターゼ(ALP)、及び総プロテイン(TP)が含まれ、このほかの情報が含まれていてもよい。また、第3判別部155が取得する生化学的パラメータには、対象者の年齢が含まれ、このほかの情報が含まれていてもよい。
一例として、Sony社のPrediction Oneを用いて第3学習済モデルを作成した。この第3学習済モデルの作成には、未治療バセドウ病患者19335例と、健常者4159例の学習用データを用いた。
AUC 0.974
さらに、対象者がバセドウ病に罹患している否かの判定について、性別による依存性を確認するために、上記学習用データを男女に分けて、教師データとテストデータのデータセットを用意し、教師データを用いて第3学習済モデル(女性)と第3学習済モデル(男性)の作成を行った。
さらに、対象者がバセドウ病に罹患している否かの判定について、遊離チロキシン(FT4)レベルへの依存性を確認するために、上記学習用データから、遊離チロキシン(FT4)レベルが5ng/dl以上の重症甲状腺中毒症を判定する第3学習済モデル(重度GD)を作成するためのテストデータと、遊離チロキシン(FT4)レベルが5ng/dl未満の軽症甲状腺中毒症を判定する第3学習済モデル(軽度GD)を作成するためのテストデータを作成した。
次に、このように構成された第3実施形態の情報処理装置100の動作について説明する。
第4実施形態では、甲状腺疾患判定装置である情報処理装置100によって、対象者の血液検査に関する情報から、甲状腺ホルモンFT3、甲状腺ホルモンFT4、FT3/FT4、アルカリホスファターゼ(ALP)、及びクレアチニン(CREA)に関する情報を、第4学習済モデルに入力し、前記対象者が罹患している疾患がバセドウ病であるか無痛性甲状腺炎であるかを鑑別する診断システム1が構築される。
第4判別部155が取得する血液学的検査に関する情報には、例えば、対象者の甲状腺ホルモンFT3、甲状腺ホルモンFT4、FT3/FT4、アルカリホスファターゼ(ALP)、及びクレアチニン(CREA)が含まれ、このほかの情報が含まれていてもよい。
一例として、Sony社のPrediction Oneを用いて第4学習済モデルを作成した。この第4学習済モデルの作成には、未治療バセドウ病患者19335例と、未治療無痛性甲状腺炎患者3267例の学習用データを用いた。
次に、このように構成された第4実施形態の情報処理装置100の動作について説明する。
第5実施形態では、甲状腺疾患判定装置である情報処理装置100によって、対象者の血液検査に関する情報から、アルカリホスファターゼ(ALP)、クレアチニン(CREA)、総プロテイン(TP)、γグルタミルトランスペプチダーゼ(γGTP)、白血球(WBC)に関する情報を、第5学習済モデルに入力し、前記対象者が罹患している疾患がバセドウ病であるか無痛性甲状腺炎であるかの鑑別する診断システム1が構築される。
第5判別部155が取得する血液学的検査に関する情報には、例えば、対象者のアルカリホスファターゼ(ALP)、クレアチニン(CREA)、総プロテイン(TP)、γグルタミルトランスペプチダーゼ(γGTP)、白血球(WBC)が含まれ、このほかの情報が含まれていてもよい。
一例として、Sony社のPrediction Oneを用いて第5学習済モデルを作成した。この第5学習済モデルの作成には、未治療バセドウ病患者19335例と、未治療無痛性甲状腺炎患者3267例の学習用データを用いた。
次に、このように構成された第5実施形態の情報処理装置100の動作について説明する。
第6実施形態では、甲状腺疾患判定装置である情報処理装置100によって、対象者の年齢と、該対象者の血液検査に関する情報から、総コレステロール(TC)、コリンエステラーゼ(ChE)、クレアチニン(CREA)、及びクレアチンホスホキナーゼ(CPK)に関する情報を、第6学習済モデルに入力し、対象者が無痛性甲状腺炎に罹患している否か判定する診断システム1が構築される。
第6判別部155が取得する血液学的検査に関する情報には、例えば、対象者の総コレステロール(TC)、コリンエステラーゼ(ChE)、クレアチニン(CREA)、及びクレアチンホスホキナーゼ(CPK)が含まれ、このほかの情報が含まれていてもよい。また、第6判別部155が取得する生化学的パラメータには、対象者の年齢が含まれ、このほかの情報が含まれていてもよい。
一例として、Sony社のPrediction Oneを用いて第6学習済モデルを作成した。この第6学習済モデルの作成には、未治療無痛性甲状腺炎患者3267例と、健常者4159例の学習用データを用いた。
次に、このように構成された第6実施形態の情報処理装置100の動作について説明する。
第7実施形態では、甲状腺疾患判定装置である情報処理装置100によって、対象者の血液検査に関する情報から、クレアチニン(CREA)、総コレステロール(TC)、コリンエステラーゼ(ChE)、クレアチンホスホキナーゼ(CPK)、及び好塩基球(Ba)に関する情報を、第7学習済モデルに入力し、対象者が甲状腺中毒症に罹患している否か判定する診断システム1が構築される。
第7判別部155が取得する血液学的検査に関する情報には、例えば、対象者のクレアチニン(CREA)、総コレステロール(TC)、コリンエステラーゼ(ChE)、クレアチンホスホキナーゼ(CPK)、及び好塩基球(Ba)が含まれ、このほかの情報が含まれていてもよい。
一例として、Sony社のPrediction Oneを用いて第7学習済モデルを作成した。この第7学習済モデルの作成には、未治療バセドウ病患者19335例と、未治療無痛性甲状腺炎患者3267例と、健常者4159例の学習用データを用いた。
AUC 0.951
次に、このように構成された第7実施形態の情報処理装置100の動作について説明する。
第1実施形態~第7実施形態に記載の甲状腺疾患判定装置による診断システム1は、独立して用いても、任意に組み合わせて用いてもよい。例えば、第1実施形態と第2実施形態との組み合わせ、第3実施形態と第7実施形態との組み合わせ、第4実施形態と第7実施形態との組み合わせが挙げられる。
第1実施例1は、第1実施形態と第2実施形態とを組み合わせた例である。実施例1の情報処理装置100は、対象者の年齢と、該対象者の血液検査に関する情報から、トータルプロテイン(TP)、クレアチニン(CREA)、好中球(Neu)、及び総ビリルビン(T-Bill)に関する情報とを、第2学習済モデルに入力し、前記対象者の甲状腺刺激ホルモン(TSH)が正常値の範囲であるか否かに関する情報を出力する第2判別部と、前記対象者の血液検査に関する情報から、トータルプロテイン(TP)、コリンエステラーゼ(ChE)、トータルコレステロール(TC)、クレアチニン(CREA)、及びクレアチンフォスフォキナーゼ(CPK)に関する情報を、第1学習済モデルに入力し、前記対象者の甲状腺刺激ホルモン(TSH)が治療を要する範囲であるか否かに関する情報を出力する第1判別部と、を有する。
第1実施例2は、第3実施形態と第7実施形態とを組み合わせた例である。実施例2の情報処理装置100は、対象者の血液検査に関する情報から、クレアチニン(CREA)、トータルコレステロール(TC)、コリンエステラーゼ(ChE)、クレアチンフォスフォキナーゼ(CPK)、及び好塩基球(Ba)に関する情報を、第7学習済モデルに入力し、前記対象者が甲状腺中毒症に罹患している否かに関する情報を出力する第7判別部と、前記対象者の年齢と、前記対象者の血液検査に関する情報から、クレアチニン(CREA)、トータルコレステロール(TC)、アルカリフォスファターゼ(ALP)、及びトータルプロテイン(TP)に関する情報とを、第3学習済モデルに入力し、前記対象者がバセドウ病に罹患している否かに関する情報を出力する第3判別部と、を有する。
第1実施例3は、第4実施形態と第7実施形態とを組み合わせた例である。実施例3の情報処理装置100は、対象者の血液検査に関する情報から、クレアチニン(CREA)、トータルコレステロール(TC)、コリンエステラーゼ(ChE)、クレアチンフォスフォキナーゼ(CPK)、及び好塩基球(Ba)に関する情報を、第7学習済モデルに入力し、前記対象者が甲状腺中毒症に罹患している否かに関する情報を出力する第7判別部と、対象者の血液検査に関する情報から、甲状腺ホルモンFT3、甲状腺ホルモンFT4、FT3/FT4、アルカリホスファターゼ(ALP)、及びクレアチニン(CREA)に関する情報を、第4学習済モデルに入力し、前記対象者が罹患している疾患がバセドウ病であるか無痛性甲状腺炎であるかの鑑別に関する情報を出力する第4判別部と、を有する。
第1実施形態~第7実施形態に記載の甲状腺疾患判定装置による診断システム1は、インターネット等の通信ネットワークを介して通信可能に接続された、視力予後判定装置であるサーバ(情報処理装置100)と利用者端末200により構成されてもよい。
Claims (7)
- 対象者の血液検査に関する情報から、総プロテイン(TP)、コリンエステラーゼ(ChE)、総コレステロール(TC)、クレアチニン(CREA)、及びクレアチンホスホキナーゼ(CPK)に関する情報を、第1学習済モデルに入力し、前記対象者の甲状腺刺激ホルモン(TSH)が治療を要する範囲であるか否かに関する情報を出力する第1判別部を有する、
情報処理装置。 - 対象者の年齢と、該対象者の血液検査に関する情報から、総プロテイン(TP)、クレアチニン(CREA)、好中球(Neu)、及び総ビリルビン(T-Bill)に関する情報とを、第2学習済モデルに入力し、前記対象者の甲状腺刺激ホルモン(TSH)が正常値の範囲であるか否かに関する情報を出力する第2判別部を有する、
情報処理装置。 - 対象者の年齢と、該対象者の血液検査に関する情報から、クレアチニン(CREA)、総コレステロール(TC)、アルカリホスファターゼ(ALP)、及び総プロテイン(TP)に関する情報とを、第3学習済モデルに入力し、前記対象者がバセドウ病に罹患している否かに関する情報を出力する第3判別部を有する、
情報処理装置。 - 対象者の血液検査に関する情報から、甲状腺ホルモンFT3、甲状腺ホルモンFT4、FT3/FT4、アルカリホスファターゼ(ALP)、及びクレアチニン(CREA)に関する情報を、第4学習済モデルに入力し、前記対象者が罹患している疾患がバセドウ病であるか無痛性甲状腺炎であるかの鑑別に関する情報を出力する第4判別部を有する、
情報処理装置。 - 対象者の血液検査に関する情報から、アルカリホスファターゼ(ALP)、クレアチニン(CREA)、総プロテイン(TP)、γグルタミルトランスペプチダーゼ(γGTP)、白血球(WBC)に関する情報を、第5学習済モデルに入力し、前記対象者が罹患している疾患がバセドウ病であるか無痛性甲状腺炎であるかの鑑別に関する情報を出力する第5判別部を有する、
情報処理装置。 - 対象者の年齢と、該対象者の血液検査に関する情報から、総コレステロール(TC)、コリンエステラーゼ(ChE)、クレアチニン(CREA)、及びクレアチンホスホキナーゼ(CPK)に関する情報を、第6学習済モデルに入力し、前記対象者が無痛性甲状腺炎に罹患している否かに関する情報を出力する第6判別部を有する、
情報処理装置。 - 対象者の血液検査に関する情報から、クレアチニン(CREA)、総コレステロール(TC)、コリンエステラーゼ(ChE)、クレアチンホスホキナーゼ(CPK)、及び好塩基球(Ba)に関する情報を、第7学習済モデルに入力し、前記対象者が甲状腺中毒症に罹患している否かに関する情報を出力する第7判別部を有する、
情報処理装置。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2023515505A JPWO2022225000A1 (ja) | 2021-04-21 | 2022-04-21 | |
CN202280029915.8A CN117616282A (zh) | 2021-04-21 | 2022-04-21 | 提供与甲状腺疾病的诊断相关的参考信息的信息处理装置 |
US18/556,849 US20240221957A1 (en) | 2021-04-21 | 2022-04-21 | Information processing device for providing reference information relating to diagnosis of thyroid disease |
EP22791771.3A EP4328590A1 (en) | 2021-04-21 | 2022-04-21 | Information processing device for providing reference information relating to diagnosis of thyroid disease |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2021072021 | 2021-04-21 | ||
JP2021-072021 | 2021-04-21 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022225000A1 true WO2022225000A1 (ja) | 2022-10-27 |
Family
ID=83722305
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2022/018395 WO2022225000A1 (ja) | 2021-04-21 | 2022-04-21 | 甲状腺疾患の診断に関する参考情報を提供する情報処理装置 |
Country Status (5)
Country | Link |
---|---|
US (1) | US20240221957A1 (ja) |
EP (1) | EP4328590A1 (ja) |
JP (1) | JPWO2022225000A1 (ja) |
CN (1) | CN117616282A (ja) |
WO (1) | WO2022225000A1 (ja) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011089923A1 (ja) * | 2010-01-22 | 2011-07-28 | ベックマン コールター, インコーポレイテッド | 自動分析装置および分析方法 |
JP2012509061A (ja) * | 2008-11-17 | 2012-04-19 | ベラサイト インコーポレイテッド | 疾患診断のための分子プロファイリングの方法および組成物 |
JP2020073885A (ja) | 2012-03-05 | 2020-05-14 | オサケユキテュア・アークティック・パートナーズ・アクチボラゲットOy Arctic Partners Ab | 前立腺癌のリスク及び前立腺容積を予測する方法及び装置 |
-
2022
- 2022-04-21 CN CN202280029915.8A patent/CN117616282A/zh active Pending
- 2022-04-21 JP JP2023515505A patent/JPWO2022225000A1/ja active Pending
- 2022-04-21 WO PCT/JP2022/018395 patent/WO2022225000A1/ja active Application Filing
- 2022-04-21 EP EP22791771.3A patent/EP4328590A1/en active Pending
- 2022-04-21 US US18/556,849 patent/US20240221957A1/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012509061A (ja) * | 2008-11-17 | 2012-04-19 | ベラサイト インコーポレイテッド | 疾患診断のための分子プロファイリングの方法および組成物 |
WO2011089923A1 (ja) * | 2010-01-22 | 2011-07-28 | ベックマン コールター, インコーポレイテッド | 自動分析装置および分析方法 |
JP2020073885A (ja) | 2012-03-05 | 2020-05-14 | オサケユキテュア・アークティック・パートナーズ・アクチボラゲットOy Arctic Partners Ab | 前立腺癌のリスク及び前立腺容積を予測する方法及び装置 |
Non-Patent Citations (4)
Title |
---|
AI YOSHIHARA, HIROSHI YOSHIMURA, NATSUKO WATANABE, MIHO FUKUSHITA, MASAKO MATSUMOTO, NAMI SUZUKI, AYAKO HOSHIYAMA, AYA KINOSHITA, : "O2-1-6 Prediction of Thyroid Toxicosis with General Bloodwork Using Machine Learning Software", NIHON NAIBUNPI GAKKAI ZASSHI - FOLIA ENDOCRINOLOGICA JAPONICA, NIHON NAIBUNPI GAKKAI, JP, vol. 97, no. 1, 1 April 2021 (2021-04-01) - 24 April 2021 (2021-04-24), JP , pages 263, XP009540431, ISSN: 0029-0661, DOI: 10.1507/endocrine.97.1_105 * |
AYAKO HOSHIYAMA, HIROSHI YOSHIMURA, RAN YOSHIMURA, KENTARO MIKURA, AYA KINOSHITA, AI SUZUKI, TAKAKO MIMATSU, NAMI SUZUKI, MIHO FUK: "O2-9-2 Prediction of Thyroid Gland Functional Deterioration with General Gloodwork Using Artificial Intelligence (AI)", FOLIA ENDOCRINOLOGICA JAPONICA, vol. 97, no. 1, 1 April 2021 (2021-04-01) - 24 April 2021 (2021-04-24), pages 269, XP009540433, DOI: 10.1507/endocrine.97.1_105 * |
NATSUKO WATANABE, HIROSHI YOSHIMURA, AI YOSHIHARA, MASAKO MATSUMOTO, MIHO FUKUSHITA, NAMI SUZUKI, AYAKO HOSHIYAMA, AI SUZUKI, TAKA: "O2-1-3 Differential Prediction Model for Graves' Disease and Painless Thyroiditis Using Artificial Intelligence (Machine Learning)", FOLIA ENDOCRINOLOGICA JAPONICA, vol. 97, no. 1, 1 April 2021 (2021-04-01) - 24 April 2021 (2021-04-24), pages 263, XP009540432, DOI: 10.1507/endocrine.97.1_105 * |
RAO AMULYA.R., RENUKA B.S.: "A Machine Learning Approach to Predict Thyroid Disease at Early Stages of Diagnosis", 2020 IEEE INTERNATIONAL CONFERENCE FOR INNOVATION IN TECHNOLOGY (INOCON), IEEE, 6 November 2020 (2020-11-06) - 8 November 2020 (2020-11-08), pages 1 - 4, XP055977943, ISBN: 978-1-7281-9744-9, DOI: 10.1109/INOCON50539.2020.9298252 * |
Also Published As
Publication number | Publication date |
---|---|
US20240221957A1 (en) | 2024-07-04 |
CN117616282A (zh) | 2024-02-27 |
JPWO2022225000A1 (ja) | 2022-10-27 |
EP4328590A1 (en) | 2024-02-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11404166B2 (en) | Systems and methods for mining of medical data | |
Chuang et al. | Interrelations of lead levels in bone, venous blood, and umbilical cord blood with exogenous lead exposure through maternal plasma lead in peripartum women. | |
US20210217528A1 (en) | Method to personalize a biomarker signal and medical procedures using the same | |
Witberg et al. | Diagnostic performance of angiogram-derived fractional flow reserve: a pooled analysis of 5 prospective cohort studies | |
Brasen et al. | Home management of warfarin treatment through a real-time supervised telemedicine solution: a randomized controlled trial | |
US20130275050A1 (en) | Methods and systems for integrated health systems | |
Otto Mattsson et al. | Patient self‐testing of white blood cell count and differentiation: A study of feasibility and measurement performance in a population of Danish cancer patients | |
Shi et al. | The utilization of retinal nerve fiber layer thickness to predict cognitive deterioration | |
Neu et al. | Non-invasive assessment of mitochondrial oxygen metabolism in the critically ill patient using the protoporphyrin IX-triplet state lifetime technique—a feasibility study | |
Boletsis et al. | Connecting the player to the doctor: utilising serious games for cognitive training & screening | |
Labrecque et al. | Functional clinical impairments and frailty in interstitial lung disease patients | |
Kier | Biostatistical applications in epidemiology | |
Fraiche et al. | Identification of need for ultrasound enhancing agent study (the IN-USE Study) | |
WO2022225000A1 (ja) | 甲状腺疾患の診断に関する参考情報を提供する情報処理装置 | |
JP6401297B2 (ja) | インスリン分泌能分析装置、当該装置を備えるインスリン分泌能分析システム及びインスリン分泌能分析方法 | |
Wood et al. | Wearable sensor-based performance status assessment in cancer: A pilot multicenter study from the Alliance for Clinical Trials in Oncology (A19_Pilot2) | |
EP3905255A1 (en) | Mapping a patient to clinical trials for patient specific clinical decision support | |
KR102536982B1 (ko) | 문진표를 이용한 인공지능 기반의 질병 예측 방법 및 장치 | |
Albu et al. | Expansion of electronic health record-based screening, prevention, and management of diabetes in New York City | |
Al Zomia et al. | Blood Parameter Profiles and Their Clinical Implications in Hypertensive Patients: A Retrospective Chart Review | |
Kartal et al. | Outpatient treatment of pulmonary embolism: sPESI score and highly sensitive troponin may prove helpful | |
Palmas | Pocket Evidence Based Medicine | |
KLOOSTERMAN | A Patient-centered educational approach to intrathoracic impedance remote monitoring can reduce hospitalizations | |
Kinikar et al. | Converging Diagnostic Paths: A Unified Approach to Acid-Base Imbalances in COPD Patients | |
Riley et al. | Comparison of Hispanics to Caucasians in Metabolic Evaluation of Nephrolithiasis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22791771 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2023515505 Country of ref document: JP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 202280029915.8 Country of ref document: CN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2022791771 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2022791771 Country of ref document: EP Effective date: 20231121 |