Metrology Standards for Quantitative Imaging Biomarkers

Abstract

Although investigators in the imaging community have been active in developing and evaluating quantitative imaging biomarkers (QIBs), the development and implementation of QIBs have been hampered by the inconsistent or incorrect use of terminology or methods for technical performance and statistical concepts. Technical performance is an assessment of how a test performs in reference objects or subjects under controlled conditions. In this article, some of the relevant statistical concepts are reviewed, methods that can be used for evaluating and comparing QIBs are described, and some of the technical performance issues related to imaging biomarkers are discussed. More consistent and correct use of terminology and study design principles will improve clinical research, advance regulatory science, and foster better care for patients who undergo imaging studies.

Figure 1:

Plot of measured phantom volumes versus true volumes. The identity line (black line) with slope equal to 1 and intercept at zero would represent a biomarker with no bias. The blue line, where the slope is equal to 1 but the intercept is not zero, represents a biomarker with fixed or constant bias (ie, not proportional to the true values). The green line, where the slope is constant but not equal to 1, represents a biomarker with proportional bias. The red line, which is curvilinear, represents a biomarker with nonconstant (nonlinear) bias. See reference for more detail.

Figure 2:

Precision profile box plot. At reference values (ie, the true value of the measurand) from 1 to 10, the mean and quartiles of the measured values are plotted. The heights of the box plots illustrate how the variance of the measured values decreases as the reference value increases.

Figure 3:

Flow diagram of the three main types of QIB performance evaluation studies. Estimates should also be accompanied by measures of uncertainty, such as 95% confidence intervals.