Format

Send to

Choose Destination
Stat Methods Med Res. 2015 Feb;24(1):68-106. doi: 10.1177/0962280214537390. Epub 2014 Jun 11.

Quantitative imaging biomarkers: a review of statistical methods for computer algorithm comparisons.

Author information

1
Cleveland Clinic Foundation, Cleveland, OH, USA obuchon@ccf.org.
2
Cornell University, Ithaca, NY, USA.
3
National Institutes of Health, Rockville, MD, USA.
4
Cleveland Clinic Foundation, Cleveland, OH, USA.
5
Elucid Bioimaging Inc., Wenham, MA, USA.
6
University of California, Los Angeles, CA, USA.
7
Duke University, Durham, NC, USA.
8
University of Wisconsin-Madison, Madison, WI, USA.
9
University of Chicago, Chicago, IL, USA.
10
Food and Drug Administration/CDRH, Silver Spring, MD, USA.
11
Biostatistics Consulting, LLC, Kensington, MD, USA.
12
MGH/Harvard Medical School, Boston, MA, USA.
13
George Washington University, NW Washington, DC, USA.
14
University of Washington, Seattle, WA, USA.
15
University of South Florida, Tampa, FL, USA.
16
H. Moffitt Cancer Center, Tampa, FL, USA.
17
Columbia University, New York, NY, USA.

Abstract

Quantitative biomarkers from medical images are becoming important tools for clinical diagnosis, staging, monitoring, treatment planning, and development of new therapies. While there is a rich history of the development of quantitative imaging biomarker (QIB) techniques, little attention has been paid to the validation and comparison of the computer algorithms that implement the QIB measurements. In this paper we provide a framework for QIB algorithm comparisons. We first review and compare various study designs, including designs with the true value (e.g. phantoms, digital reference images, and zero-change studies), designs with a reference standard (e.g. studies testing equivalence with a reference standard), and designs without a reference standard (e.g. agreement studies and studies of algorithm precision). The statistical methods for comparing QIB algorithms are then presented for various study types using both aggregate and disaggregate approaches. We propose a series of steps for establishing the performance of a QIB algorithm, identify limitations in the current statistical literature, and suggest future directions for research.

KEYWORDS:

agreement; bias; image metrics; imaging biomarkers; precision; quantitative imaging; repeatability; reproducibility

PMID:
24919829
PMCID:
PMC4263694
DOI:
10.1177/0962280214537390
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Atypon Icon for PubMed Central
Loading ...
Support Center