使用 的 Amazon Comprehend Medical 範例 AWS CLI - AWS SDK 程式碼範例
動作

文件 AWS SDK AWS 範例 SDK 儲存庫中有更多可用的 GitHub 範例。

本文為英文版的機器翻譯版本，如內容有任何歧義或不一致之處，概以英文版為準。

使用 的 Amazon Comprehend Medical 範例 AWS CLI

下列程式碼範例示範如何搭配 Amazon Comprehend Medical AWS Command Line Interface 使用 來執行動作和實作常見案例。

Actions 是大型程式的程式碼摘錄，必須在內容中執行。雖然 動作會示範如何呼叫個別服務函數，但您可以在其相關案例中查看內容中的動作。

每個範例都包含完整原始程式碼的連結，您可以在其中找到如何在內容中設定和執行程式碼的指示。

主題

動作

下列程式碼範例示範如何使用 describe-entities-detection-v2-job。

AWS CLI

描述實體偵測任務

下列describe-entities-detection-v2-job範例顯示與非同步實體偵測工作相關聯的屬性。

aws comprehendmedical describe-entities-detection-v2-job \
    --job-id "ab9887877365fe70299089371c043b96"

輸出：

{
    "ComprehendMedicalAsyncJobProperties": {
        "JobId": "ab9887877365fe70299089371c043b96",
        "JobStatus": "COMPLETED",
        "SubmitTime": "2020-03-18T21:20:15.614000+00:00",
        "EndTime": "2020-03-18T21:27:07.350000+00:00",
        "ExpirationTime": "2020-07-16T21:20:15+00:00",
        "InputDataConfig": {
            "S3Bucket": "comp-med-input",
            "S3Key": ""
        },
        "OutputDataConfig": {
            "S3Bucket": "comp-med-output",
            "S3Key": "867139942017-EntitiesDetection-ab9887877365fe70299089371c043b96/"
        },
        "LanguageCode": "en",
        "DataAccessRoleArn": "arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole",
        "ModelVersion": "DetectEntitiesModelV20190930"
    }
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的批次APIs。

下列程式碼範例示範如何使用 describe-icd10-cm-inference-job。

AWS CLI

描述 ICD-10-CM 推論任務

下列describe-icd10-cm-inference-job範例說明具有指定任務 ID 的請求推論任務屬性。

aws comprehendmedical describe-icd10-cm-inference-job \
    --job-id "5780034166536cdb52ffa3295a1b00a7"

輸出：

{
    "ComprehendMedicalAsyncJobProperties": {
        "JobId": "5780034166536cdb52ffa3295a1b00a7",
        "JobStatus": "COMPLETED",
        "SubmitTime": "2020-05-18T21:20:15.614000+00:00",
        "EndTime": "2020-05-18T21:27:07.350000+00:00",
        "ExpirationTime": "2020-09-16T21:20:15+00:00",
        "InputDataConfig": {
            "S3Bucket": "comp-med-input",
            "S3Key": "AKIAIOSFODNN7EXAMPLE"
        },
        "OutputDataConfig": {
            "S3Bucket": "comp-med-output",
            "S3Key": "AKIAIOSFODNN7EXAMPLE"
        },
        "LanguageCode": "en",
        "DataAccessRoleArn": "arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole",
        "ModelVersion":  "0.1.0"
    }
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Ontology 連結批次分析。

下列程式碼範例示範如何使用 describe-phi-detection-job。

AWS CLI

描述 PHI 偵測任務

下列describe-phi-detection-job範例顯示與非同步受保護健康資訊 (PHI) 偵測工作相關聯的屬性。

aws comprehendmedical describe-phi-detection-job \
    --job-id "4750034166536cdb52ffa3295a1b00a3"

輸出：

{
    "ComprehendMedicalAsyncJobProperties": {
        "JobId": "4750034166536cdb52ffa3295a1b00a3",
        "JobStatus": "COMPLETED",
        "SubmitTime": "2020-03-19T20:38:37.594000+00:00",
        "EndTime": "2020-03-19T20:45:07.894000+00:00",
        "ExpirationTime": "2020-07-17T20:38:37+00:00",
        "InputDataConfig": {
            "S3Bucket": "comp-med-input",
            "S3Key": ""
        },
        "OutputDataConfig": {
            "S3Bucket": "comp-med-output",
            "S3Key": "867139942017-PHIDetection-4750034166536cdb52ffa3295a1b00a3/"
        },
        "LanguageCode": "en",
        "DataAccessRoleArn": "arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole",
        "ModelVersion": "PHIModelV20190903"
    }
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的批次APIs。

下列程式碼範例示範如何使用 describe-rx-norm-inference-job。

AWS CLI

描述 a RxNorm 推論任務

下列describe-rx-norm-inference-job範例說明具有指定任務 ID 的請求推論任務屬性。

aws comprehendmedical describe-rx-norm-inference-job \
    --job-id "eg8199877365fc70299089371c043b96"

輸出：

{
    "ComprehendMedicalAsyncJobProperties": {
        "JobId": "g8199877365fc70299089371c043b96",
        "JobStatus": "COMPLETED",
        "SubmitTime": "2020-05-18T21:20:15.614000+00:00",
        "EndTime": "2020-05-18T21:27:07.350000+00:00",
        "ExpirationTime": "2020-09-16T21:20:15+00:00",
        "InputDataConfig": {
            "S3Bucket": "comp-med-input",
            "S3Key": "AKIAIOSFODNN7EXAMPLE"
        },
        "OutputDataConfig": {
            "S3Bucket": "comp-med-output",
            "S3Key": "AKIAIOSFODNN7EXAMPLE"
        },
        "LanguageCode": "en",
        "DataAccessRoleArn": "arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole",
        "ModelVersion": "0.0.0"
    }
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Ontology 連結批次分析。

下列程式碼範例示範如何使用 describe-snomedct-inference-job。

AWS CLI

描述 SNOMED CT 推論任務

下列describe-snomedct-inference-job範例說明具有指定任務 ID 的請求推論任務屬性。

aws comprehendmedical describe-snomedct-inference-job \
    --job-id "2630034166536cdb52ffa3295a1b00a7"

輸出：

{
    "ComprehendMedicalAsyncJobProperties": {
        "JobId": "2630034166536cdb52ffa3295a1b00a7",
        "JobStatus": "COMPLETED",
        "SubmitTime": "2021-12-18T21:20:15.614000+00:00",
        "EndTime": "2021-12-18T21:27:07.350000+00:00",
        "ExpirationTime": "2022-05-16T21:20:15+00:00",
        "InputDataConfig": {
            "S3Bucket": "comp-med-input",
            "S3Key": "AKIAIOSFODNN7EXAMPLE"
        },
        "OutputDataConfig": {
            "S3Bucket": "comp-med-output",
            "S3Key": "AKIAIOSFODNN7EXAMPLE"
        },
        "LanguageCode": "en",
        "DataAccessRoleArn": "arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole",
        "ModelVersion":  "0.1.0"
    }
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Ontology 連結批次分析。

下列程式碼範例示範如何使用 detect-entities-v2。

AWS CLI

範例 1：直接從文字偵測實體

下列detect-entities-v2範例顯示偵測到的實體，並直接從輸入文字根據類型對其進行標籤。

aws comprehendmedical detect-entities-v2 \
    --text "Sleeping trouble on present dosage of Clonidine. Severe rash on face and leg, slightly itchy."

輸出：

{
    "Id": 0,
    "BeginOffset": 38,
    "EndOffset": 47,
    "Score": 0.9942955374717712,
    "Text": "Clonidine",
    "Category": "MEDICATION",
    "Type": "GENERIC_NAME",
    "Traits": []
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的偵測實體版本 2。

範例 2：從檔案路徑偵測實體

下列detect-entities-v2範例顯示偵測到的實體，並根據檔案路徑中的類型對其進行標籤。

aws comprehendmedical detect-entities-v2 \
    --text file://medical_entities.txt

medical_entities.txt 的內容：

{
    "Sleeping trouble on present dosage of Clonidine. Severe rash on face and leg, slightly itchy."
}

輸出：

{
    "Id": 0,
    "BeginOffset": 38,
    "EndOffset": 47,
    "Score": 0.9942955374717712,
    "Text": "Clonidine",
    "Category": "MEDICATION",
    "Type": "GENERIC_NAME",
    "Traits": []
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的偵測實體版本 2。

  • 如需 API 詳細資訊，請參閱 AWS CLI 命令參考中的 DetectEntitiesV2。

下列程式碼範例示範如何使用 detect-phi。

AWS CLI

範例 1：直接從文字偵測受保護的健康資訊 (PHI)

下列detect-phi範例會直接從輸入文字顯示偵測到的受保護健康資訊 (PHI) 實體。

aws comprehendmedical detect-phi \
    --text "Patient Carlos Salazar presented with rash on his upper extremities and dry cough. He lives at 100 Main Street, Anytown, USA where he works from his home as a carpenter."

輸出：

{
    "Entities": [
        {
            "Id": 0,
            "BeginOffset": 8,
            "EndOffset": 21,
            "Score": 0.9914507269859314,
            "Text": "Carlos Salazar",
            "Category": "PROTECTED_HEALTH_INFORMATION",
            "Type": "NAME",
            "Traits": []
        },
        {
            "Id": 1,
            "BeginOffset": 94,
            "EndOffset": 109,
            "Score": 0.871849775314331,
            "Text": "100 Main Street, Anytown, USA",
            "Category": "PROTECTED_HEALTH_INFORMATION",
            "Type": "ADDRESS",
            "Traits": []
        },
        {
            "Id": 2,
            "BeginOffset": 145,
            "EndOffset": 154,
            "Score": 0.8302185535430908,
            "Text": "carpenter",
            "Category": "PROTECTED_HEALTH_INFORMATION",
            "Type": "PROFESSION",
            "Traits": []
        }
    ],
    "ModelVersion": "0.0.0"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Detect PHI。

範例 2：直接從檔案路徑偵測保護運作狀態資訊 (PHI)

下列detect-phi範例顯示從檔案路徑偵測到的受保護健康資訊 (PHI) 實體。

aws comprehendmedical detect-phi \
    --text file://phi.txt

phi.txt 的內容：

"Patient Carlos Salazar presented with a rash on his upper extremities and a dry cough. He lives at 100 Main Street, Anytown, USA, where he works from his home as a carpenter."

輸出：

{
    "Entities": [
        {
            "Id": 0,
            "BeginOffset": 8,
            "EndOffset": 21,
            "Score": 0.9914507269859314,
            "Text": "Carlos Salazar",
            "Category": "PROTECTED_HEALTH_INFORMATION",
            "Type": "NAME",
            "Traits": []
        },
        {
            "Id": 1,
            "BeginOffset": 94,
            "EndOffset": 109,
            "Score": 0.871849775314331,
            "Text": "100 Main Street, Anytown, USA",
            "Category": "PROTECTED_HEALTH_INFORMATION",
            "Type": "ADDRESS",
            "Traits": []
        },
        {
            "Id": 2,
            "BeginOffset": 145,
            "EndOffset": 154,
            "Score": 0.8302185535430908,
            "Text": "carpenter",
            "Category": "PROTECTED_HEALTH_INFORMATION",
            "Type": "PROFESSION",
            "Traits": []
        }
    ],
    "ModelVersion": "0.0.0"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Detect PHI。

  • 如需 API 詳細資訊，請參閱 AWS CLI 命令參考中的 DetectPhi。

下列程式碼範例示範如何使用 infer-icd10-cm。

AWS CLI

範例 1：直接從文字偵測醫療狀況實體和 ICD-10-CM Ontology 的連結

下列infer-icd10-cm範例會標記偵測到的醫療狀況實體，並將這些實體與 2019 年國際疾病分類臨床修改 (ICD-10-CM) 版中的代碼連結。

aws comprehendmedical infer-icd10-cm \
    --text "The patient complains of abdominal pain, has a long-standing history of diabetes treated with Micronase daily."

輸出：

{
    "Entities": [
        {
            "Id": 0,
            "Text": "abdominal pain",
            "Category": "MEDICAL_CONDITION",
            "Type": "DX_NAME",
            "Score": 0.9475538730621338,
            "BeginOffset": 28,
            "EndOffset": 42,
            "Attributes": [],
            "Traits": [
                {
                    "Name": "SYMPTOM",
                    "Score": 0.6724207401275635
                }
            ],
            "ICD10CMConcepts": [
                {
                    "Description": "Unspecified abdominal pain",
                    "Code": "R10.9",
                    "Score": 0.6904221177101135
                },
                {
                    "Description": "Epigastric pain",
                    "Code": "R10.13",
                    "Score": 0.1364113688468933
                },
                {
                    "Description": "Generalized abdominal pain",
                    "Code": "R10.84",
                    "Score": 0.12508003413677216
                },
                {
                    "Description": "Left lower quadrant pain",
                    "Code": "R10.32",
                    "Score": 0.10063883662223816
                },
                {
                    "Description": "Lower abdominal pain, unspecified",
                    "Code": "R10.30",
                    "Score": 0.09933677315711975
                }
            ]
        },
        {
            "Id": 1,
            "Text": "diabetes",
            "Category": "MEDICAL_CONDITION",
            "Type": "DX_NAME",
            "Score": 0.9899052977561951,
            "BeginOffset": 75,
            "EndOffset": 83,
            "Attributes": [],
            "Traits": [
                {
                    "Name": "DIAGNOSIS",
                    "Score": 0.9258432388305664
                }
            ],
            "ICD10CMConcepts": [
                {
                    "Description": "Type 2 diabetes mellitus without complications",
                    "Code": "E11.9",
                    "Score": 0.7158446311950684
                },
                {
                    "Description": "Family history of diabetes mellitus",
                    "Code": "Z83.3",
                    "Score": 0.5704703330993652
                },
                {
                    "Description": "Family history of other endocrine, nutritional and metabolic diseases",
                    "Code": "Z83.49",
                    "Score": 0.19856023788452148
                },
                {
                    "Description": "Type 1 diabetes mellitus with ketoacidosis without coma",
                    "Code": "E10.10",
                    "Score": 0.13285516202449799
                },
                {
                    "Description": "Type 2 diabetes mellitus with hyperglycemia",
                    "Code": "E11.65",
                    "Score": 0.0993388369679451
                }
            ]
        }
    ],
    "ModelVersion": "0.1.0"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的推論 ICD10-CM。

範例 2：從檔案路徑偵測醫療狀況實體和 ICD-10-CM Ontology 的連結

下列infer-icd-10-cm範例會標記偵測到的醫療狀況實體，並將這些實體與 2019 年國際疾病臨床修改分類 (ICD-10-CM) 版中的代碼連結。

aws comprehendmedical infer-icd10-cm \
    --text file://icd10cm.txt

icd10cm.txt 的內容：

{
    "The patient complains of abdominal pain, has a long-standing history of diabetes treated with Micronase daily."
}

輸出：

{
    "Entities": [
        {
            "Id": 0,
            "Text": "abdominal pain",
            "Category": "MEDICAL_CONDITION",
            "Type": "DX_NAME",
            "Score": 0.9475538730621338,
            "BeginOffset": 28,
            "EndOffset": 42,
            "Attributes": [],
            "Traits": [
                {
                    "Name": "SYMPTOM",
                    "Score": 0.6724207401275635
                }
            ],
            "ICD10CMConcepts": [
                {
                    "Description": "Unspecified abdominal pain",
                    "Code": "R10.9",
                    "Score": 0.6904221177101135
                },
                {
                    "Description": "Epigastric pain",
                    "Code": "R10.13",
                    "Score": 0.1364113688468933
                },
                {
                    "Description": "Generalized abdominal pain",
                    "Code": "R10.84",
                    "Score": 0.12508003413677216
                },
                {
                    "Description": "Left lower quadrant pain",
                    "Code": "R10.32",
                    "Score": 0.10063883662223816
                },
                {
                    "Description": "Lower abdominal pain, unspecified",
                    "Code": "R10.30",
                    "Score": 0.09933677315711975
                }
            ]
        },
        {
            "Id": 1,
            "Text": "diabetes",
            "Category": "MEDICAL_CONDITION",
            "Type": "DX_NAME",
            "Score": 0.9899052977561951,
            "BeginOffset": 75,
            "EndOffset": 83,
            "Attributes": [],
            "Traits": [
                {
                    "Name": "DIAGNOSIS",
                    "Score": 0.9258432388305664
                }
            ],
            "ICD10CMConcepts": [
                {
                    "Description": "Type 2 diabetes mellitus without complications",
                    "Code": "E11.9",
                    "Score": 0.7158446311950684
                },
                {
                    "Description": "Family history of diabetes mellitus",
                    "Code": "Z83.3",
                    "Score": 0.5704703330993652
                },
                {
                    "Description": "Family history of other endocrine, nutritional and metabolic diseases",
                    "Code": "Z83.49",
                    "Score": 0.19856023788452148
                },
                {
                    "Description": "Type 1 diabetes mellitus with ketoacidosis without coma",
                    "Code": "E10.10",
                    "Score": 0.13285516202449799
                },
                {
                    "Description": "Type 2 diabetes mellitus with hyperglycemia",
                    "Code": "E11.65",
                    "Score": 0.0993388369679451
                }
            ]
        }
    ],
    "ModelVersion": "0.1.0"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Infer-ICD10-CM。

  • 如需 API 詳細資訊，請參閱 AWS CLI 命令參考中的 InferIcd10Cm。

下列程式碼範例示範如何使用 infer-rx-norm。

AWS CLI

範例 1：直接從文字偵測藥品實體和連結至 RxNorm

下列infer-rx-norm範例顯示並標記偵測到的藥品實體，並將這些實體連結至 National Library of MedicineCUI 資料庫的概念識別碼 (Rx RxNorm )。

aws comprehendmedical infer-rx-norm \
    --text "Patient reports taking Levothyroxine 125 micrograms p.o. once daily, but denies taking Synthroid."

輸出：

{
    "Entities": [
        {
            "Id": 0,
            "Text": "Levothyroxine",
            "Category": "MEDICATION",
            "Type": "GENERIC_NAME",
            "Score": 0.9996285438537598,
            "BeginOffset": 23,
            "EndOffset": 36,
            "Attributes": [
                {
                    "Type": "DOSAGE",
                    "Score": 0.9892290830612183,
                    "RelationshipScore": 0.9997978806495667,
                    "Id": 1,
                    "BeginOffset": 37,
                    "EndOffset": 51,
                    "Text": "125 micrograms",
                    "Traits": []
                },
                {
                    "Type": "ROUTE_OR_MODE",
                    "Score": 0.9988924860954285,
                    "RelationshipScore": 0.998291552066803,
                    "Id": 2,
                    "BeginOffset": 52,
                    "EndOffset": 56,
                    "Text": "p.o.",
                    "Traits": []
                },
                {
                    "Type": "FREQUENCY",
                    "Score": 0.9953463673591614,
                    "RelationshipScore": 0.9999889135360718,
                    "Id": 3,
                    "BeginOffset": 57,
                    "EndOffset": 67,
                    "Text": "once daily",
                    "Traits": []
                }
            ],
            "Traits": [],
            "RxNormConcepts": [
                {
                    "Description": "Levothyroxine Sodium 0.125 MG Oral Tablet",
                    "Code": "966224",
                    "Score": 0.9912070631980896
                },
                {
                    "Description": "Levothyroxine Sodium 0.125 MG Oral Capsule",
                    "Code": "966405",
                    "Score": 0.8698278665542603
                },
                {
                    "Description": "Levothyroxine Sodium 0.125 MG Oral Tablet [Synthroid]",
                    "Code": "966191",
                    "Score": 0.7448257803916931
                },
                {
                    "Description": "levothyroxine",
                    "Code": "10582",
                    "Score": 0.7050482630729675
                },
                {
                    "Description": "Levothyroxine Sodium 0.125 MG Oral Tablet [Levoxyl]",
                    "Code": "966190",
                    "Score": 0.6921631693840027
                }
            ]
        },
        {
            "Id": 4,
            "Text": "Synthroid",
            "Category": "MEDICATION",
            "Type": "BRAND_NAME",
            "Score": 0.9946461319923401,
            "BeginOffset": 86,
            "EndOffset": 95,
            "Attributes": [],
            "Traits": [
                {
                    "Name": "NEGATION",
                    "Score": 0.5167351961135864
                }
            ],
            "RxNormConcepts": [
                {
                    "Description": "Synthroid",
                    "Code": "224920",
                    "Score": 0.9462039470672607
                },
                {
                    "Description": "Levothyroxine Sodium 0.088 MG Oral Tablet [Synthroid]",
                    "Code": "966282",
                    "Score": 0.8309829235076904
                },
                {
                    "Description": "Levothyroxine Sodium 0.125 MG Oral Tablet [Synthroid]",
                    "Code": "966191",
                    "Score": 0.4945160448551178
                },
                {
                    "Description": "Levothyroxine Sodium 0.05 MG Oral Tablet [Synthroid]",
                    "Code": "966247",
                    "Score": 0.3674522042274475
                },
                {
                    "Description": "Levothyroxine Sodium 0.025 MG Oral Tablet [Synthroid]",
                    "Code": "966158",
                    "Score": 0.2588822841644287
                }
            ]
        }
    ],
    "ModelVersion": "0.0.0"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Infer RxNorm。

範例 2：從檔案路徑偵測藥品實體和連結至 RxNorm 。

下列infer-rx-norm範例顯示並標記偵測到的藥品實體，並將這些實體連結至 National Library of MedicineCUI 資料庫的概念識別碼 (Rx RxNorm )。

aws comprehendmedical infer-rx-norm \
    --text file://rxnorm.txt

rxnorm.txt 的內容：

{
    "Patient reports taking Levothyroxine 125 micrograms p.o. once daily, but denies taking Synthroid."
}

輸出：

{
    "Entities": [
        {
            "Id": 0,
            "Text": "Levothyroxine",
            "Category": "MEDICATION",
            "Type": "GENERIC_NAME",
            "Score": 0.9996285438537598,
            "BeginOffset": 23,
            "EndOffset": 36,
            "Attributes": [
                {
                    "Type": "DOSAGE",
                    "Score": 0.9892290830612183,
                    "RelationshipScore": 0.9997978806495667,
                    "Id": 1,
                    "BeginOffset": 37,
                    "EndOffset": 51,
                    "Text": "125 micrograms",
                    "Traits": []
                },
                {
                    "Type": "ROUTE_OR_MODE",
                    "Score": 0.9988924860954285,
                    "RelationshipScore": 0.998291552066803,
                    "Id": 2,
                    "BeginOffset": 52,
                    "EndOffset": 56,
                    "Text": "p.o.",
                    "Traits": []
                },
                {
                    "Type": "FREQUENCY",
                    "Score": 0.9953463673591614,
                    "RelationshipScore": 0.9999889135360718,
                    "Id": 3,
                    "BeginOffset": 57,
                    "EndOffset": 67,
                    "Text": "once daily",
                    "Traits": []
                }
            ],
            "Traits": [],
            "RxNormConcepts": [
                {
                    "Description": "Levothyroxine Sodium 0.125 MG Oral Tablet",
                    "Code": "966224",
                    "Score": 0.9912070631980896
                },
                {
                    "Description": "Levothyroxine Sodium 0.125 MG Oral Capsule",
                    "Code": "966405",
                    "Score": 0.8698278665542603
                },
                {
                    "Description": "Levothyroxine Sodium 0.125 MG Oral Tablet [Synthroid]",
                    "Code": "966191",
                    "Score": 0.7448257803916931
                },
                {
                    "Description": "levothyroxine",
                    "Code": "10582",
                    "Score": 0.7050482630729675
                },
                {
                    "Description": "Levothyroxine Sodium 0.125 MG Oral Tablet [Levoxyl]",
                    "Code": "966190",
                    "Score": 0.6921631693840027
                }
            ]
        },
        {
            "Id": 4,
            "Text": "Synthroid",
            "Category": "MEDICATION",
            "Type": "BRAND_NAME",
            "Score": 0.9946461319923401,
            "BeginOffset": 86,
            "EndOffset": 95,
            "Attributes": [],
            "Traits": [
                {
                    "Name": "NEGATION",
                    "Score": 0.5167351961135864
                }
            ],
            "RxNormConcepts": [
                {
                    "Description": "Synthroid",
                    "Code": "224920",
                    "Score": 0.9462039470672607
                },
                {
                    "Description": "Levothyroxine Sodium 0.088 MG Oral Tablet [Synthroid]",
                    "Code": "966282",
                    "Score": 0.8309829235076904
                },
                {
                    "Description": "Levothyroxine Sodium 0.125 MG Oral Tablet [Synthroid]",
                    "Code": "966191",
                    "Score": 0.4945160448551178
                },
                {
                    "Description": "Levothyroxine Sodium 0.05 MG Oral Tablet [Synthroid]",
                    "Code": "966247",
                    "Score": 0.3674522042274475
                },
                {
                    "Description": "Levothyroxine Sodium 0.025 MG Oral Tablet [Synthroid]",
                    "Code": "966158",
                    "Score": 0.2588822841644287
                }
            ]
        }
    ],
    "ModelVersion": "0.0.0"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Infer RxNorm。

  • 如需 API 詳細資訊，請參閱 AWS CLI 命令參考中的 InferRxNorm。

下列程式碼範例示範如何使用 infer-snomedct。

AWS CLI

範例：直接從文字偵測實體並連結至 SNOMED CT Ontology

下列infer-snomedct範例示範如何偵測醫療實體，並將其連結至 2021-03 版本的 Systematized Nomenclature of Medicine， Clinical Terms (SNOMED CT) 概念。

aws comprehendmedical infer-snomedct \
    --text "The patient complains of abdominal pain, has a long-standing history of diabetes treated with Micronase daily."

輸出：

{
    "Entities": [
        {
            "Id": 3,
            "BeginOffset": 26,
            "EndOffset": 40,
            "Score": 0.9598260521888733,
            "Text": "abdominal pain",
            "Category": "MEDICAL_CONDITION",
            "Type": "DX_NAME",
            "Traits": [
                {
                    "Name": "SYMPTOM",
                    "Score": 0.6819021701812744
                }
            ]
        },
        {
            "Id": 4,
            "BeginOffset": 73,
            "EndOffset": 81,
            "Score": 0.9905840158462524,
            "Text": "diabetes",
            "Category": "MEDICAL_CONDITION",
            "Type": "DX_NAME",
            "Traits": [
                {
                    "Name": "DIAGNOSIS",
                    "Score": 0.9255214333534241
                }
            ]
        },
        {
            "Id": 1,
            "BeginOffset": 95,
            "EndOffset": 104,
            "Score": 0.6371926665306091,
            "Text": "Micronase",
            "Category": "MEDICATION",
            "Type": "BRAND_NAME",
            "Traits": [],
            "Attributes": [
                {
                    "Type": "FREQUENCY",
                    "Score": 0.9761165380477905,
                    "RelationshipScore": 0.9984188079833984,
                    "RelationshipType": "FREQUENCY",
                    "Id": 2,
                    "BeginOffset": 105,
                    "EndOffset": 110,
                    "Text": "daily",
                    "Category": "MEDICATION",
                    "Traits": []
                }
            ]
        }
    ],
    "UnmappedAttributes": [],
    "ModelVersion": "1.0.0"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 InferSNOMEDCT。

  • 如需 API 詳細資訊，請參閱 AWS CLI 命令參考中的 InferSnomedct。

下列程式碼範例示範如何使用 list-entities-detection-v2-jobs。

AWS CLI

若要列出實體偵測任務

下列list-entities-detection-v2-jobs範例列出目前的非同步偵測任務。

aws comprehendmedical list-entities-detection-v2-jobs

輸出：

{
    "ComprehendMedicalAsyncJobPropertiesList": [
        {
            "JobId": "ab9887877365fe70299089371c043b96",
            "JobStatus": "COMPLETED",
            "SubmitTime": "2020-03-19T20:38:37.594000+00:00",
            "EndTime": "2020-03-19T20:45:07.894000+00:00",
            "ExpirationTime": "2020-07-17T20:38:37+00:00",
            "InputDataConfig": {
                "S3Bucket": "comp-med-input",
                "S3Key": ""
            },
            "OutputDataConfig": {
                "S3Bucket": "comp-med-output",
                "S3Key": "867139942017-EntitiesDetection-ab9887877365fe70299089371c043b96/"
            },
            "LanguageCode": "en",
            "DataAccessRoleArn": "arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole",
            "ModelVersion": "DetectEntitiesModelV20190930"
        }
    ]
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的批次APIs。

下列程式碼範例示範如何使用 list-icd10-cm-inference-jobs。

AWS CLI

若要列出所有目前的 ICD-10-CM 推論任務

下列範例顯示 list-icd10-cm-inference-jobs操作如何傳回目前非同步 ICD-10-CM 批次推論任務的清單。

aws comprehendmedical list-icd10-cm-inference-jobs

輸出：

{
    "ComprehendMedicalAsyncJobPropertiesList": [
        {
            "JobId": "5780034166536cdb52ffa3295a1b00a7",
            "JobStatus": "COMPLETED",
            "SubmitTime": "2020-05-19T20:38:37.594000+00:00",
            "EndTime": "2020-05-19T20:45:07.894000+00:00",
            "ExpirationTime": "2020-09-17T20:38:37+00:00",
            "InputDataConfig": {
                "S3Bucket": "comp-med-input",
                "S3Key": "AKIAIOSFODNN7EXAMPLE"
            },
            "OutputDataConfig": {
                "S3Bucket": "comp-med-output",
                "S3Key": "AKIAIOSFODNN7EXAMPLE"
            },
            "LanguageCode": "en",
            "DataAccessRoleArn": "arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole",
            "ModelVersion":  "0.1.0"
        }
    ]
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Ontology 連結批次分析。

下列程式碼範例示範如何使用 list-phi-detection-jobs。

AWS CLI

若要列出受保護的健康資訊 (PHI) 偵測任務

下列list-phi-detection-jobs範例列出目前受保護的健康資訊 (PHI) 偵測任務

aws comprehendmedical list-phi-detection-jobs

輸出：

{
    "ComprehendMedicalAsyncJobPropertiesList": [
        {
            "JobId": "4750034166536cdb52ffa3295a1b00a3",
            "JobStatus": "COMPLETED",
            "SubmitTime": "2020-03-19T20:38:37.594000+00:00",
            "EndTime": "2020-03-19T20:45:07.894000+00:00",
            "ExpirationTime": "2020-07-17T20:38:37+00:00",
            "InputDataConfig": {
                "S3Bucket": "comp-med-input",
                "S3Key": ""
            },
            "OutputDataConfig": {
                "S3Bucket": "comp-med-output",
                "S3Key": "867139942017-PHIDetection-4750034166536cdb52ffa3295a1b00a3/"
            },
            "LanguageCode": "en",
            "DataAccessRoleArn": "arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole",
            "ModelVersion": "PHIModelV20190903"
        }
    ]
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的批次APIs。

下列程式碼範例示範如何使用 list-rx-norm-inference-jobs。

AWS CLI

若要列出所有目前的 Rx-Norm 推論任務

下列範例顯示如何list-rx-norm-inference-jobs傳回目前非同步 Rx-Norm 批次推論任務的清單。

aws comprehendmedical list-rx-norm-inference-jobs

輸出：

{
    "ComprehendMedicalAsyncJobPropertiesList": [
        {
            "JobId": "4980034166536cfb52gga3295a1b00a3",
            "JobStatus": "COMPLETED",
            "SubmitTime": "2020-05-19T20:38:37.594000+00:00",
            "EndTime": "2020-05-19T20:45:07.894000+00:00",
            "ExpirationTime": "2020-09-17T20:38:37+00:00",
            "InputDataConfig": {
                "S3Bucket": "comp-med-input",
                "S3Key": "AKIAIOSFODNN7EXAMPLE"
            },
            "OutputDataConfig": {
                "S3Bucket": "comp-med-output",
                "S3Key": "AKIAIOSFODNN7EXAMPLE"
            },
            "LanguageCode": "en",
            "DataAccessRoleArn": "arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole",
            "ModelVersion": "0.0.0"
        }
    ]
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Ontology 連結批次分析。

下列程式碼範例示範如何使用 list-snomedct-inference-jobs。

AWS CLI

若要列出所有 SNOMED CT 推論任務

下列範例顯示 list-snomedct-inference-jobs操作如何傳回目前非同步 SNOMED CT 批次推論任務的清單。

aws comprehendmedical list-snomedct-inference-jobs

輸出：

{
    "ComprehendMedicalAsyncJobPropertiesList": [
        {
            "JobId": "5780034166536cdb52ffa3295a1b00a7",
            "JobStatus": "COMPLETED",
            "SubmitTime": "2020-05-19T20:38:37.594000+00:00",
            "EndTime": "2020-05-19T20:45:07.894000+00:00",
            "ExpirationTime": "2020-09-17T20:38:37+00:00",
            "InputDataConfig": {
                "S3Bucket": "comp-med-input",
                "S3Key": "AKIAIOSFODNN7EXAMPLE"
            },
            "OutputDataConfig": {
                "S3Bucket": "comp-med-output",
                "S3Key": "AKIAIOSFODNN7EXAMPLE"
            },
            "LanguageCode": "en",
            "DataAccessRoleArn": "arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole",
            "ModelVersion":  "0.1.0"
        }
    ]
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Ontology 連結批次分析。

下列程式碼範例示範如何使用 start-entities-detection-v2-job。

AWS CLI

啟動實體偵測任務

下列start-entities-detection-v2-job範例會啟動非同步實體偵測任務。

aws comprehendmedical start-entities-detection-v2-job \
    --input-data-config "S3Bucket=comp-med-input" \
    --output-data-config "S3Bucket=comp-med-output" \
    --data-access-role-arn arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole \
    --language-code en

輸出：

{
    "JobId": "ab9887877365fe70299089371c043b96"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的批次APIs。

下列程式碼範例示範如何使用 start-icd10-cm-inference-job。

AWS CLI

若要啟動 ICD-10-CM 推論任務

下列start-icd10-cm-inference-job範例會啟動 ICD-10-CM 推論批次分析任務。

aws comprehendmedical start-icd10-cm-inference-job \
    --input-data-config "S3Bucket=comp-med-input" \
    --output-data-config "S3Bucket=comp-med-output" \
    --data-access-role-arn arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole \
    --language-code en

輸出：

{
    "JobId": "ef7289877365fc70299089371c043b96"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Ontology 連結批次分析。

下列程式碼範例示範如何使用 start-phi-detection-job。

AWS CLI

啟動 PHI 偵測任務

下列start-phi-detection-job範例會啟動非同步 PHI 實體偵測任務。

aws comprehendmedical start-phi-detection-job \
    --input-data-config "S3Bucket=comp-med-input" \
    --output-data-config "S3Bucket=comp-med-output" \
    --data-access-role-arn arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole \
    --language-code en

輸出：

{
    "JobId": "ab9887877365fe70299089371c043b96"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的批次APIs。

下列程式碼範例示範如何使用 start-rx-norm-inference-job。

AWS CLI

啟動 a RxNorm 推論任務

下列start-rx-norm-inference-job範例會啟動 RxNorm 推論批次分析任務。

aws comprehendmedical start-rx-norm-inference-job \
    --input-data-config "S3Bucket=comp-med-input" \
    --output-data-config "S3Bucket=comp-med-output" \
    --data-access-role-arn arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole \
    --language-code en

輸出：

{
    "JobId": "eg8199877365fc70299089371c043b96"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Ontology 連結批次分析。

下列程式碼範例示範如何使用 start-snomedct-inference-job。

AWS CLI

啟動 SNOMED CT 推論任務

下列start-snomedct-inference-job範例會啟動 SNOMED CT 推論批次分析任務。

aws comprehendmedical start-snomedct-inference-job \
    --input-data-config "S3Bucket=comp-med-input" \
    --output-data-config "S3Bucket=comp-med-output" \
    --data-access-role-arn arn:aws:iam::867139942017:role/ComprehendMedicalBatchProcessingRole \
    --language-code en

輸出：

{
    "JobId": "dg7289877365fc70299089371c043b96"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Ontology 連結批次分析。

下列程式碼範例示範如何使用 stop-entities-detection-v2-job。

AWS CLI

停止實體偵測任務

下列stop-entities-detection-v2-job範例會停止非同步實體偵測工作。

aws comprehendmedical stop-entities-detection-v2-job \
    --job-id "ab9887877365fe70299089371c043b96"

輸出：

{
    "JobId": "ab9887877365fe70299089371c043b96"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的批次APIs。

下列程式碼範例示範如何使用 stop-icd10-cm-inference-job。

AWS CLI

停止 ICD-10-CM 推論任務

下列stop-icd10-cm-inference-job範例會停止 ICD-10-CM 推論批次分析工作。

aws comprehendmedical stop-icd10-cm-inference-job \
    --job-id "4750034166536cdb52ffa3295a1b00a3"

輸出：

{
    "JobId": "ef7289877365fc70299089371c043b96",
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Ontology 連結批次分析。

下列程式碼範例示範如何使用 stop-phi-detection-job。

AWS CLI

若要停止受保護的健康資訊 (PHI) 偵測工作

下列stop-phi-detection-job範例會停止非同步受保護健康資訊 (PHI) 偵測工作。

aws comprehendmedical stop-phi-detection-job \
    --job-id "4750034166536cdb52ffa3295a1b00a3"

輸出：

{
    "JobId": "ab9887877365fe70299089371c043b96"
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的批次APIs。

下列程式碼範例示範如何使用 stop-rx-norm-inference-job。

AWS CLI

若要停止 a RxNorm 推論任務

下列stop-rx-norm-inference-job範例會停止 ICD-10-CM 推論批次分析工作。

aws comprehendmedical stop-rx-norm-inference-job \
    --job-id "eg8199877365fc70299089371c043b96"

輸出：

{
    "JobId": "eg8199877365fc70299089371c043b96",
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Ontology 連結批次分析。

下列程式碼範例示範如何使用 stop-snomedct-inference-job。

AWS CLI

若要停止 SNOMED CT 推論任務

下列stop-snomedct-inference-job範例會停止 SNOMED CT 推論批次分析工作。

aws comprehendmedical stop-snomedct-inference-job \
    --job-id "8750034166436cdb52ffa3295a1b00a1"

輸出：

{
    "JobId": "8750034166436cdb52ffa3295a1b00a1",
}

如需詳細資訊，請參閱 Amazon Comprehend Medical 開發人員指南中的 Ontology 連結批次分析。