DCMQI

DCMQI converts imaging research formats to standard DICOM representations to enable standardized encoding and sharing of PET/CT quantitative imaging analysis results for quantitative imaging biomarker development.

Key Features:

• Format conversion and standards-based representation: Conversion routines create standards-based DICOM representations of imaging analysis results and related clinical data.
• DICOM modeling: Leverages DICOM components to model various data types generated during quantitative imaging analysis.
• DICOM object support: Uses DICOM Real World Value Mapping, Segmentation, and Structured Reporting objects for compliant representation of PET/CT QI results.
• API and developer toolkit: Provides an Application Programming Interface and conversion routines for programmatic creation of DICOM-encoded analysis objects.
• DCMTK enhancements: Introduces new API abstractions to DCMTK to simplify DICOM encoding tasks and support consistent object generation.
• SUV normalization: Implements normalization of Standardized Uptake Value (SUV) as part of quantitative PET/CT analysis.
• Tumor segmentation: Supports manual and semi-automatic tumor segmentation approaches.
• Reference region segmentation: Performs automatic segmentation of reference regions used in analysis.
• Volumetric measurements: Extracts volumetric measurements from segmentations for quantitative assessment.
• Interoperability validation: Encoded objects are validated for use in systems supporting the DICOM standard.
• Data deposition: Resulting datasets have been deposited in the QIN-HEADNECK collection of The Cancer Imaging Archive (TCIA).

Scientific Applications:

• PET/CT quantitative imaging biomarker development: Enables standardized representation and exchange of PET/CT QI analysis outputs for biomarker research.
• Assessment of treatment response in head and neck cancer (HNC): Supports workflows for evaluating treatment response in HNC using SUV normalization, segmentation, and volumetric measurements.

Methodology:

Normalization of Standardized Uptake Value (SUV); tumor segmentation using manual and semi-automatic approaches; automatic segmentation of reference regions; extraction of volumetric measurements from segmentations; conversion routines producing DICOM Real World Value Mapping, Segmentation, and Structured Reporting objects; and DCMTK API abstraction enhancements for DICOM encoding.