What the Data for This Category Looks Like
Data sources for this category include the national medical insurance information platform, hospital HIS and medical insurance interfaces, medical insurance catalog and coding library, and cross-provincial medical treatment settlement platform. The update frequency is daily.
Daily processed documents include medical insurance settlement lists, expense details, medical record homepages, prescriptions, rejection lists, and settlement reconciliation statements. Core fields involve diagnosis and treatment item codes, drug and medical supplies codes, expense amounts, insured person information, and medical institution information. Verification requires matching against standardized medical insurance catalog codes.
How These Characteristics Impact the Workflow Orchestration Link
Dispersed data sources with daily updates require workflows to support multi-data source connection and scheduled pulling, to ensure the timeliness of audit data. Diverse document types and high field standardization requirements mean multiple nodes must be configured in orchestration to handle parsing and field mapping for different documents, ensuring unified data formats. Processing volume for single or bulk settlement data fluctuates widely. Adjustable node concurrency parameters must be set to adapt to processing needs for different scenarios. The reconciliation link for rejection lists and settlement reconciliation statements requires precise field matching. A field consistency check node must be configured in the workflow to avoid audit errors caused by code mismatches.
Configuration Recommendations
Reference only; verify against the target FastGPT release and deployment documentation.
| Configuration Item | Recommended Practice | Rationale |
|---|---|---|
| Trigger Mode | Trigger via scheduled task, set interval to 1–24 hours | Matches the daily data update rhythm, covers processing needs for daily bulk settlement data |
| variables | Configure as a variable group containing settlement batch number and data source identifier | Adapts to multi-institution, multi-batch bulk audit requirements, supports passing custom parameters during API calls |
| Knowledge Base Selection | Reference the medical insurance catalog knowledge base variable defined in variables | Ensures the latest medical insurance coding library is used during audits, aligns with data update frequency requirements |
| Node Concurrency | Set to 2–5 concurrent instances | Balances processing efficiency and system load, adapts to fluctuations in single or bulk settlement data processing |
| Field Validation Rules | Match diagnosis and treatment item, drug and medical supplies codes against the medical insurance catalog | Ensures audit data complies with standardization requirements, reduces errors caused by code mismatches |
| Failure Retry Configuration | Retry 2 times after failure, set interval to 30 seconds | Addresses temporary interface fluctuations or data synchronization delays, ensures process stability |
Three Common Configuration Mistakes
- Phenomenon: When configuring a knowledge base search node in the workflow, after referencing the knowledge base parameter passed via
variables, medical insurance catalog data cannot be matched. Cause: The unique identifier of the knowledge base is not configured invariables, or the passed parameter does not match the knowledge base identifier in the system. - Phenomenon: A timeout error occurs during workflow runtime, with status code 504. Cause: No reasonable node timeout time is set, or the concurrency number is too high, causing processing delay to exceed the system's carrying threshold.
- Phenomenon: Manually added audio and video tags cannot be displayed normally in the reply node. Cause: The rich text rendering configuration of the workflow is not enabled, or the tag format does not comply with the specifications supported by the system.
How to Confirm the Configuration Is Correct
- Trigger a test run of the workflow, check whether the parameters configured in
variablesare correctly loaded in the node logs. - Check the matching results of the knowledge base search node, confirm whether code verification data related to the medical insurance catalog is returned.
- View the node configuration of the workflow, confirm whether the
nodeIdof each associated node matches correctly, to ensure the data flow path is error-free. - Simulate a process run for a single settlement data, verify whether the field verification node can correctly identify abnormal codes or mismatched fields.
- View the failure retry records of the workflow, confirm whether the preset retry mechanism is triggered in temporary abnormal scenarios.
The values provided here are common starting points, and should be evaluated against the reader's own samples.
Question material comes from public community discussions. Configuration values are common starting points and should be measured against your own samples. Verified on 2026-09-28.