Importance of Statistical Metrology Framework for Quantitative Imaging Applications*

Quantitative imaging is being increasingly used to diagnose disease, predict patients’ outcomes, and monitor and adapt treatment. In clinical trials, quantitative imaging often serves as a non-invasive endpoint which can provide results earlier than traditional patient outcomes.

The precision and bias of quantitative imaging measurements drive their utility at the bedside and in clinical trials. Precision describes the closeness of replicate measurements, while bias describes how the measurements differ from the true value [1]. These technical performance characteristics are important in order to understand patients’ measurements, as well as to plan and analyze clinical trials [2].

In this presentation we examine several clinical trials using quantitative imaging. We discuss how the technical performance characteristics of the imaging biomarkers are used to determine sample size, identify eligible subjects, estimate treatment effect, and interpret measurements and change in measurements over time.

[1] Kessler LG, Barnhart HX, Buckler AJ, et al. The emerging science of quantitative imaging biomarkers: terminology and definitions for scientific studies and for regulatory submissions. SMMR 2015; 24: 9-26.
[2] Obuchowski NA, Mozley DP, Matthews D, Buckler A, Bullen J, and Jackson E. Statistical
Considerations for Planning Clinical Trials with Quantitative Imaging Biomarkers. Journal National Cancer Institute 2019; 111: 19-26.