Metadata Quality in Imaging Trials: A Knowledge Graph-Based Approach
View abstract on PubMed
Summary
This summary is machine-generated.We developed a method to convert DICOM metadata into a knowledge graph, enabling advanced analytics and automated quality assessments for imaging trials. This facilitates robust data quality checks, crucial for complex research studies.
Area Of Science
- Medical Imaging Informatics
- Data Science
- Clinical Trials
Background
- Digital Imaging and Communications in Medicine (DICOM) is the standard for medical imaging.
- Managing and analyzing DICOM metadata is complex, especially in large-scale imaging trials.
- Ensuring metadata quality is critical for the validity of clinical trial results.
Purpose Of The Study
- To present a novel method for transforming DICOM metadata into a structured knowledge graph.
- To enable advanced analytical capabilities for DICOM metadata.
- To facilitate automated metadata quality assessment in imaging trials.
Main Methods
- Development of a transformation pipeline to convert DICOM metadata into a knowledge graph format.
- Implementation of custom metrics for evaluating metadata quality attributes, such as completeness.
- Application of the knowledge graph for automated quality control procedures.
Main Results
- Successful transformation of DICOM metadata into a queryable knowledge graph.
- Demonstration of metadata quality assessment using defined custom metrics.
- Validation of the method's utility for automated quality checks in imaging trials.
Conclusions
- The proposed method provides a robust framework for leveraging DICOM metadata through knowledge graphs.
- Automated quality assessment of metadata is achievable, enhancing data integrity in clinical research.
- This approach offers practical benefits for managing complex, multifaceted imaging trials.