Extract Adverse Drug Events (ADE)

**Extract Adverse Drug Events (ADE):**

This model is engineered for the extraction of adverse drug events (ADEs) from unstructured clinical texts, leveraging several components finely tuned for this purpose:
- Entity Recognition: Initially, the model accurately identifies entities related to adverse events (such as rash, nausea) and drug mentions within the text.
- Assertion Status Detection: Subsequently, it assigns an assertion status (e.g., present, negated, historical, hypothetical) to each identified ADE entity, taking into account the surrounding context.
- Document Classification: The model then classifies the entire document, discerning whether it contains a report of an ADE. This classification aids in filtering documents more likely to possess relevant ADE information.
- Relation Extraction: The final step involves the detection of relationships between the extracted ADE entities and drug entities, thereby identifying pairs of medication and associated adverse events.

The model addresses the critical need to identify adverse drug reactions (ADRs) - unintended and sometimes harmful responses to medications. It's crucial for healthcare professionals and patients to understand the potential side effects for informed care and decision-making. Especially in cases with multiple drug mentions, this model proficiently correlates drugs with their respective adverse reactions, discerning if an event is drug-induced.

In the digital era, patients frequently share medication experiences across various platforms like online reviews, social media, and forums. This model meticulously scans such diverse content, including clinical notes, to detect drug-associated adverse reactions. It serves as a proactive tool for early ADR identification, thereby enhancing patient safety and refining drug prescription practices.

**Additional Model Information**
- [Full model info on John Snow Labs Models Hub](https://nlp.johnsnowlabs.com/2023/06/17/explain_clinical_doc_ade_en.html)
- **Domain:** Clinical Text Analysis
- **Subdomain:** Pharmacovigilance and Medication Information
- **Predictable entities:** DRUG, ADE
- **Assertion status entities:** absent, present, conditional, associated_with_someone_else, hypothetical, possible.
- **Predicted relations between entities:** 0, 1
- **Deployment Identifier:** 26. Extract Adverse Drug Events (ADE)

**How to run this model:**
1. Acquire a John Snow Labs Pay As You Go (PAYG) license from [John Snow Labs](https://my.johnsnowlabs.com/).
2. Import this listing
3. Use the attached notebook to deploy the model with **26. Extract Adverse Drug Events (ADE)** as the model parameter. **Do not use the Open button on this page which appears after importing this listing. It will fail to deploy a model and does not work yet, you must use the attached notebook.**

This model comes with optimized CPU and GPU builds. You can select which one to deploy via the notebook.

**How to obtain a PAYG license for John Snow Labs models:**

1. Access [my.JohnSnowLabs.com](https://my.johnsnowlabs.com/) and log in to your account. If you don't have an account, create one. 
2. Go to the Get License page. 
3. Switch to the PAYG Subscription tab and provide your credit card details. 
4. Carefully review the End User License Agreement and the Terms and Conditions documents. If you agree, click on the Create Subscription button. 
5. Once the process is complete, you will find your PAY-As-You-GO license listed on the My Subscriptions page. 
6. Visit the My Subscriptions page and copy the PAYG license key by clicking on the copy icon in the License Key column.
7. Go to your Databricks notebook and paste your JSL-license into the JSL-License field in the top of the notebook. You are now ready to go!

**提取药物不良事件（ADE，Adverse Drug Events）：**

本模型专为从非结构化临床文本中提取药物不良事件（ADE）而开发，针对该任务精细调校了多个核心组件：
- 实体识别：首先，模型可精准识别文本中与不良事件相关的实体（如皮疹、恶心）以及药物提及内容。
- 断言状态检测：随后，结合上下文信息，为每个识别出的ADE实体分配断言状态（如存在、否定、既往、假设性）。
- 文档分类：随后对整篇文档进行分类，判断其是否包含ADE相关报告，该分类步骤有助于筛选出更可能携带有效ADE信息的文档。
- 关系抽取：最后一步为检测提取出的ADE实体与药物实体之间的关联关系，从而识别出药物与对应不良事件的配对关系。

本模型解决了识别药物不良反应（ADR，Adverse Drug Reactions）——即药物引发的意外且有时具有危害性的反应——的关键需求。对于医护人员和患者而言，了解药物潜在副作用以做出知情的诊疗决策至关重要。尤其是在存在多种药物提及的场景中，本模型可熟练关联药物与其对应的不良反应，辨别某一不良事件是否由药物诱发。

在数字化时代，患者常于各类平台分享用药体验，如在线评论、社交媒体及论坛。本模型可细致扫描包括临床笔记在内的多样化内容，以检测与药物相关的不良反应。它可作为早期ADR识别的前瞻性工具，进而提升患者安全水平并优化药物处方实践。

**附加模型信息**
- [约翰·斯诺实验室模型中心完整模型信息](https://nlp.johnsnowlabs.com/2023/06/17/explain_clinical_doc_ade_en.html)
- **领域**：临床文本分析
- **子领域**：药物警戒与用药信息
- **可预测实体**：药物（DRUG）、药物不良事件（ADE）
- **断言状态实体**：缺失、存在、有条件、与他人相关、假设性、可能
- **实体间预测关系**：0、1
- **部署标识符**：26. 提取药物不良事件（ADE）

**如何运行此模型：**
1. 从[约翰·斯诺实验室](https://my.johnsnowlabs.com/)获取约翰·斯诺实验室随用随付（PAYG，Pay As You Go）许可证。
2. 导入此模型条目
3. 使用附带的笔记本部署模型，将**26. 提取药物不良事件（ADE）**作为模型参数。**请勿使用导入此条目后页面上出现的“打开”按钮，该按钮目前无法成功部署模型，必须使用附带的笔记本。**

本模型提供了优化后的CPU与GPU构建版本，可通过笔记本选择部署版本。

**如何获取约翰·斯诺实验室模型的随用随付许可证：**
1. 访问[my.JohnSnowLabs.com](https://my.johnsnowlabs.com/)并登录账号，若无账号则先注册。
2. 进入获取许可证页面。
3. 切换至随用随付订阅标签页，提供您的信用卡信息。
4. 仔细审阅最终用户许可协议与条款文档，若同意则点击“创建订阅”按钮。
5. 流程完成后，您可在“我的订阅”页面找到您的随用随付许可证。
6. 进入“我的订阅”页面，点击许可证密钥列的复制图标，复制您的PAYG许可证密钥。
7. 进入您的Databricks笔记本，在笔记本顶部的JSL-License字段中粘贴您的JSL许可证密钥，即可完成配置！

搜集汇总

背景与挑战

背景概述

该模型通过实体识别、断言状态判断、文档分类和关系提取四步流程，从临床文本中检测药物与不良事件的关联关系，支持药物安全监测和用药决策。其适用于社交媒体、临床记录等多源文本分析，需通过John Snow Labs平台获取许可后部署使用。