Criteria2Query

Criteria2Query 2.0 is published! Online Demo

Introduction

Criteria2Query (C2Q) is an automatic cohort identification system. It enhances human-computer collaboration to convert complex eligibility criteria text into more accurate and feasible cohort SQL queries. It synergizes machine efficiency and human intelligence of domain experts to enable real-time user intervention for criteria selection and simplification, parsing error correction, and context-dependent concept mapping.

Features

• An editable user interface with functions to prioritize or simplify the eligibility criteria text for cohort querying;
• Accessible and portable cohort SQL query formulation based on the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) version 5;
• Real-time cohort query execution with result visualization.

Interface and use case example

System Requirements

• Java 8+
• Apache Maven 3
• Apache Tomcat
• Python 3.7.6+
• PostgreSQL DBMS (to demonstrate the real-time cohort SQL query execution, not strictly required)

Dependencies

• SynPUF_1K and SynPUF_5% datasets in CDM Version 5.2.2 format.
• OMOP CDM Vocabulary version 5 files. These can be obtained from Athena.
• These are for the demonstration of real-time cohort SQL query execution (not strictly required)

Getting Started

1. Download and install everything based on the system requirements above.

2. Git clone this repository.

3. Download the negation scope detection model and move it to the folder NegationDetection.

4. Create a virtual environment in Python and install packages based on venv_requirements.txt. (Instruction: https://packaging.python.org/en/latest/guides/installing-using-pip-and-virtual-environments/)

5. Change the directories of Negation Detection and the Python virtual environment in the file /criteria2query/src/main/java/edu/columbia/dbmi/ohdsims/pojo/GlobalSetting.java

   //Change the directories (examples)
   public final static String negateDetectionFolder = "/opt/tomcat/NegationDetection";
   public final static String virtualEnvFolder = "/opt/tomcat/python_virtualenvs/C2Q_NEGATION/bin"; // or "D:\\C2Q\\python_virtualenvs\\C2Q_NEGATION\\Scripts";

6. Import SynPUF_1K and SynPUF_5% datasets to your PostgreSQL DBMS (You can skip this step if they are already imported.)

   • Download the SynPUF_1K and SynPUF_5% datasets in CDM Version 5.2.2 format.
   • Download the OMOP CDM Vocabulary version 5 files from Athena.
   • Follow the instruction here (https://github.com/OHDSI/CommonDataModel/tree/v5.2.2/PostgreSQL) to create your instantiations of the Common Data Model for SynPUF_1K and SynPUF_5%, respectively.

7. Connect to your own database (SynPUF_1K and SynPUF_5%) by changing the URL, user, and password in the file /criteria2query/src/main/java/edu/columbia/dbmi/ohdsims/pojo/GlobalSetting.java

   //Connect to the databases
   public final static String databaseURL1K = "jdbc:postgresql://localhost/synpuf1k";
   public final static String databaseURL5pct = "jdbc:postgresql://localhost/synpuf5pct";
   public final static String databaseUser = "Please connect to a database.";
   public final static String databasePassword = "*****";

8. Deploy C2Q and visit it in your web browser.

Publications

Fang, Y., Idnay, B., Sun, Y., Liu, H., Chen, Z., Marder, K., Xu, H., Schnall, R., & Weng, C. (2022). Combining human and machine intelligence for clinical trial eligibility querying. Journal of the American Medical Informatics Association : JAMIA, ocac051. Advance online publication. https://doi.org/10.1093/jamia/ocac051

Yuan, C., Ryan, P. B., Ta, C., Guo, Y., Li, Z., Hardin, J., Makadia, R., Jin, P., Shang, N., Kang, T., & Weng, C. (2019). Criteria2Query: a natural language interface to clinical databases for cohort definition. Journal of the American Medical Informatics Association : JAMIA, 26(4), 294–305. https://doi.org/10.1093/jamia/ocy178

Support

If you have any questions/comments/feedback, please submit a form here or contact Dr. Chunhua Weng at Columbia University.