Secondary Capture

A Secondary Capture Image is a DICOM SOP Class for images converted from a non-DICOM format or captured from a video signal. It serves as a container for digitized film, screen captures, scanned documents, or imported JPEGs. The primary purpose is to integrate these external images into a PACS or VNA workflow, ensuring they can be stored, queried, and retrieved using standard DICOM services, even though they lack the rich acquisition parameters of a primary modality like CT or MR.

The defining characteristic of a Secondary Capture instance is the absence of detailed technical acquisition data. Unlike a primary CT Image Storage SOP Class, which contains tags for kVp, exposure time, and slice thickness, a Secondary Capture object typically lacks these modality-specific attributes. The DICOM Tag (0008,0060) Modality is set to 'OT' (Other) or a specific secondary capture value, explicitly signaling that the image is a converted representation, not raw acquisition data.

The conversion process involves rendering a non-DICOM image into a DICOM-compliant pixel data format. This often requires specifying a Transfer Syntax for encoding, such as JPEG lossy compression. The original pixel data is encapsulated, and critical DICOM header information, like the Patient Name (0010,0010) and Study Instance UID, is manually or automatically populated. The result is a valid DICOM Part 10 file, but the pixel data is a facsimile of the original, not the raw detector output.

Secondary Capture SOP Classes exist for both single-frame and multi-frame images. Multi-frame Secondary Capture is used for converting video loops, such as an ultrasound cine clip exported from a non-DICOM system, into a single DICOM object. Each frame of the video becomes a frame in the DICOM instance, allowing the entire dynamic sequence to be managed as one entity within a PACS, preserving the temporal relationship between frames.

A significant risk with Secondary Capture images is the presence of Burned-in Annotations. Since the image is a screen capture or a scan of a printout, patient demographics or hospital identifiers may be permanently rendered into the pixel data itself. This creates a major challenge for DICOM De-identification, as standard header anonymization is insufficient. Automated optical character recognition (OCR) and image masking are required to remove this Protected Health Information (PHI) before secondary use.

Secondary Capture plays a crucial role in integrating non-DICOM devices into a digital workflow. For example, a surgical microscope that outputs only an HDMI video signal can be connected to a frame grabber. The capture software packages the video stream into Secondary Capture Image Storage SOP Instances and uses STOW-RS or DIMSE C-STORE to send them to the PACS. This allows the images to be associated with the correct patient study via the Modality Worklist, even though the microscope itself has no DICOM capability.

The fundamental unit of DICOM interoperability, defined as the union of a specific Information Object Definition (IOD) and a DIMSE Service Group. Secondary Capture is a specific SOP Class (UID 1.2.840.10008.5.1.4.1.1.7) that defines the structure for converted images. During Association Negotiation, systems must agree on this SOP Class to successfully exchange secondary capture instances.

Defines how DICOM data is serialized into a byte stream. Secondary Capture images often use JPEG Lossy or JPEG Lossless Transfer Syntaxes for compression. The choice of Transfer Syntax is critical during conversion:

• Explicit VR Little Endian: Uncompressed, universal baseline.
• JPEG Baseline (1.2.840.10008.1.2.4.50): Common for color video captures.
• JPEG-LS Lossless: Used when diagnostic fidelity must be preserved.

A mandatory document from every medical device vendor detailing exactly which SOP Classes are supported. When integrating a Secondary Capture workflow, the conformance statement reveals:

• Whether the device acts as an SCU (sending captures) or SCP (receiving them).
• Which Transfer Syntaxes are negotiated.
• Whether Burned-in Annotations are recognized and handled.

Encodes clinical observations as machine-readable text with coded concepts rather than free-text. While Secondary Capture stores pixel data, a DICOM SR object can reference that image as evidence for a specific finding. This linkage is essential for building auditable diagnostic records where a captured ultrasound frame supports a structured measurement.

Standardized rules from DICOM Part 15 for removing Protected Health Information (PHI). Secondary Capture images are high-risk for containing PHI in Burned-in Annotations rendered into the pixel data itself. The de-identification process must:

• Strip header tags like Patient Name (0010,0010).
• Detect and redact text burned into the image raster.
• Clean private tags that may contain identifying information.

A Look-Up Table that transforms stored pixel values into standard units (e.g., Hounsfield Units for CT). A key limitation of Secondary Capture is that this transformation is often already applied and the original raw data is lost. The Modality LUT in a Secondary Capture IOD is typically an identity transform, meaning the pixel data is presentation-ready but not suitable for quantitative analysis.