Use AI Agents for Research with Bring Your Own Data

You can use AI Agents for Research to analyze your organization's datasets with Oracle Health Real-World Data (RWD).

Before using a Bring Your Own Data (BYOD) dataset, upload it to the Oracle Cloud Infrastructure (OCI) Object Storage bucket configured for Life Science AI Data Platform (AI Data Platform). Data is uploaded to Object Storage and not directly to the AI Client. BYOD data cannot be joined to RWD unless the approved tokenization, de-identification, and governance workflow has been completed. Common BYOD file types include cohort definitions, concept mappings, CTMS data, Clinical Trial protocols or other structured third party datasets acquired or licensed for approved research use. Contact AI Data Platform Support for more information.

To use AI Agents for Research to analyze BYOD:

1. Upload the dataset to the configured OCI Object Storage bucket. Verify that your file is in PDF or CSV format and that the file size is 5 GB or less. See Uploading an Object Storage Object to a Bucket for instructions on how to upload a dataset to OCI Object Storage.
2. Open your approved AI client and connect to the AI Data Platform connector or application.
3. Ask AI Agents for Research to profile, query, or analyze the uploaded dataset. Identify the dataset by name in your request.

See the prompt below for an example on how to use AI Agents to analyze your own datasets.

Using my uploaded claims dataset [dataset_name] to understand how many GLP-1 users discontinued therapy. Discontinuation is defined as no GLP-1 claim within 90 days after the member’s last GLP-1 claim. Summarize aggregate discontinuation counts by medication.