Research
Questions of reproducibility
Our proposed research asks how to tell whether a result reflects the phenomenon under study, the measurement method or the sample chosen.
Define the comparison
A study should state its question, the population to which it applies and the outcome to be measured. Those choices determine what the result can reasonably support.
Repeat the measurement
Repeat scans can test measurement consistency. Reanalysing the same image with a different method asks a separate question about the influence of analysis choices.
Examine the exceptions
Where do methods disagree, and do those cases share an acquisition setting or image characteristic? Studying exceptions could help define the limits of a finding.