HTTP Interface and External Systems for Medical Imaging

Data Characteristics

Medical imaging devices generate data for clinical trial pre-screening primarily in DICOM (Digital Imaging and Communications in Medicine) standard format. This includes medical images like CT, MRI, and X-ray. Hospital PACS (Picture Archiving and Communication System) systems typically generate and store this data. Data update frequency varies from several times daily to several times weekly, depending on trial enrollment speed and imaging schedules. DICOM files have complex hierarchical structures, containing patient information, study information, series information, and image pixel data. Beyond standard DICOM tags, custom tags specific to study protocols may also be present. Units follow medical imaging conventions, such as millimeters (mm) for dimensions and Hounsfield Units (HU) for CT density.

Constraints Imposed by Data Characteristics on "HTTP Interface and External Systems"

The complex structure and large file size of DICOM data challenge HTTP interface transmission efficiency, especially with limited network bandwidth. Patient privacy information in the data requires strict authentication and authorization mechanisms for compliance during data transfer. DICOM tags are highly standardized, but custom tags necessitate flexible field parsing capabilities in the interface to accommodate study-specific data. Imaging data often transmits as binary streams, requiring the interface to support large block uploads and downloads. The non-real-time nature of data updates means interfaces should not rely on immediate responses; asynchronous processing and status query mechanisms are more appropriate.

Configuration Guidelines

Configuration Item	Recommended Value	Rationale
`MAX_PAYLOAD_SIZE_MB`	`1024 MB`	Imaging files are large, requiring support for high-capacity single transfers.
`API_TIMEOUT_SECONDS`	`300 seconds`	Allows sufficient interface response time, considering large file uploads and network latency.
`AUTH_HEADER_NAME`	`X-API-Key`	Uses a standard custom HTTP header for API key authentication, enhancing security.
`ASYNC_UPLOAD_ENDPOINT`	`/api/v1/dicom/upload/async`	Differentiates synchronous and asynchronous uploads to handle time-consuming imaging data processing.
`DICOM_TAG_PARSING_RULES`	Calibrate based on actual measurements	Defines parsing rules according to custom DICOM tags in specific study protocols.

Common Pitfalls

Symptom: API call returns a 413 Payload Too Large error. Reason: MAX_PAYLOAD_SIZE_MB is configured too small to accommodate a single DICOM file transfer.
Symptom: External system uploads a file but remains unresponsive for a long time or returns a 504 Gateway Timeout. Reason: API_TIMEOUT_SECONDS is set too low, causing timeouts during processing or transfer of large imaging data.
Symptom: Uploaded imaging data cannot correctly identify patient information or study ID in FastGPT. Reason: DICOM_TAG_PARSING_RULES is not configured correctly, preventing parsing of custom DICOM tags in specific study protocols.

Verification Steps

Upload a typical DICOM file package. Check if the interface returns a 200 OK status code and records a file ID.
Use FastGPT's internal query function with the returned file ID. Verify that key fields like patient basic information and study number match the original DICOM data.
Upload an imaging file containing custom DICOM tags. Ensure FastGPT correctly extracts and displays this custom information, verifying field names and values match expectations.

The values provided are common starting points and should be measured 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 2026-09-21.