Integrating Multimodal Radiation Therapy Data into i2b2

Area of Science:

• Medical Informatics
• Oncology Research
• Translational Bioinformatics

Background:

• Clinical data warehouses are crucial for research, with the Informatics for Integrating Biology and the Bedside (i2b2) platform being a standard.
• Integrating radiation therapy data into i2b2 is essential for advanced cancer research and personalized treatment planning but is often unaddressed.

Purpose of the Study:

• To assess the feasibility of integrating radiation oncology data into the i2b2 platform for rectal cancer patient research.
• To enable the use of comprehensive patient data, including radiation therapy details, within the i2b2 framework.

Main Methods:

• A hybrid storage approach was implemented in an i2b2 instance, combining "Entity-Attribute-Value" with "JavaScript Object Notation (JSON)" document storage.
• Raw radiation therapy data, including dose details and dose-volume histogram (DVH) curves, were extracted using reverse engineering and vendor APIs.
• Validation involved replicating vendor screens and using R software to display DVH curves.

Main Results:

• A pipeline was developed and evaluated on 262 patients, successfully integrating radiation therapy data into the i2b2 instance.
• Preliminary use cases demonstrated the ability to fetch and visualize DVH curve data from the integrated clinical data warehouse.
• The system facilitated the analysis of radiation therapy response in cancer patients.

Conclusions:

• The integration of radiation therapy data into i2b2 enhances the platform's utility for cancer research.
• New ontology-based modules can be created using stored radiation therapy data, providing researchers with a broader spectrum of clinical information.
• This approach supports personalized treatment planning and deeper analysis of radiation therapy outcomes.