CT Volumetry Biomarker Ctte
Jump to navigation
Jump to search
This Wiki is for collaborative creation of QIBA's Volumetric CT Technical Committee materials and ongoing activities.
Mission: Investigate the technical feasibility and medical value of quantifying longitudinal changes in volume with CT.
Initial Objectives:
- Characterize the precision and accuracy of volumetric tumor measurements.
- This will be an essential prelude for understanding the threshold that will be needed to classify longitudinal changes in tumor volume as medically meaningful surrogates for changes in health status.
- Compare the sensitivity of volumetric measurements to RECIST/ World Health Organization (WHO) (conventional & modified) outcome measures.
- This will be necessary in order to determine if Progressive Disease can be detected significantly sooner with volumetric techniques than with uni- or bi-dimensional line-lengths placed on a single slice.
To accomplish this, we adopt the following flow which represents the structure of our effort as well as the resulting output we achieve:
Project Snapshot
Quantitative Response Metrics for Lung Cancer was selected as a useful concrete focus to get started.
Profiling Activities
Write Profile documents to which equipment and sites can conform to accomplish Quantification goals.
Groundwork Activities
Validate key concepts and answer key questions needed to write the Profiles and demonstrate their effectiveness.
- VolCT - Group 1A - Analyze Bias/Variance where Ground Truth is Known Deterministically
- VolCT - Group 1B - Focus on Volume Change Analysis of Pulmonary Nodules in Diagnostic Settings
- VolCT - Group 1C - Model Sources of Variability from a Systems Engineering Analysis
- VolCT - Group 2 - Set Standards for using Volumetric Imaging in Clinical Trials
Working Documents
- Initial Target Selection - describing the early prioritization discussions and settling on Lung Tumor Volumes
- Sources of Variability - Andrew Buckler
- Validation Roadmap - P. David Mozley, MD
- Process Groundwork & Profiling - Kevin O'Donnell - See also the Wiki Version