Difference between revisions of "QIBA for Statisticians"

From QIBA Wiki
Jump to: navigation, search
Line 16: Line 16:
 
* '''provide unbiased estimates''' of treatment effects when QIBs are endpoints
 
* '''provide unbiased estimates''' of treatment effects when QIBs are endpoints
  
<First Steps>
+
 
* Who to contact to get involved - qiba@rsna.org (provide staff with a list of "responder" volunteers for more in depth)
+
First Steps
* Document to start with?
+
* Review basic [[QIBA Concepts]]
* ???
+
* Review specific [[Profiles]]
 +
* Review [[Claim Guidance]] and [[Assessment Procedure Guidance]], which provide statistically-oriented background information on QIBA Profile concepts
 +
* Contact [mailto:qiba@rsna.org] to get involved

Revision as of 04:28, 19 November 2019

QIBA Mission: Improve the value and practicality of quantitative imaging biomarkers by reducing variability across devices, sites, patients, and time.

QIBA develops Profile documents that standardize methods for creating a specific biomarker with predictable performance (accuracy and reproducibility).


Statisticians benefit from QIBA by:

  • having good estimates of the technical performance of QIBs obtained using a given QIBA Profile
  • understanding the role of technical performance characteristics (repeatability, reproducibility, bias, linearity) in the study design and analysis of clinical trials with quantitative imaging biomarkers (QIBs)
  • understanding the role of technical performance characteristics in quantifying change in individual patients


Statisticians may also:

  • support testing of actors’ conformance
  • optimize study design in clinical trials with QIBs
  • determine sample size for clinical trials using QIBs
  • provide unbiased estimates of treatment effects when QIBs are endpoints


First Steps