Variance stabilization is an important step in the statistical assessment of quantitative imaging biomarkers. The objective of this study is to compare the Log and the Box–Cox transformations for variance stabilization in the context of assessing the performance of a particular quantitative imaging biomarker, the estimation of lung nodule volume from computed tomography images. First, a model is developed to generate and characterize repeated measurements typically observed in computed tomography lung nodule volume estimation. Given this model, we derive the parameter of the Box–Cox transformation that stabilizes the variance of the measurements across lung nodule volumes. Second, simulated, phantom, and clinical datasets are used to compare the Log and the Box–Cox transformations. Two metrics are used for quantifying the stability of the measurements across the transformed lung nodule volumes: the coefficient of variation for the standard deviation and the repeatability coefficient. The results for simulated, phantom, and clinical datasets show that the Box–Cox transformation generally had better variance stabilization performance compared to the Log transformation for lung nodule volume estimates from computed tomography scans.
SullivanDCObuchowskiNAKesslerLG, et al.
Metrology standards for quantitative imaging biomarkers.Radiology2015;
277: 813–825.
2.
RaunigDLMcShaneLMPennelloG, et al.
Quantitative imaging biomarkers: A review of statistical methods for technical performance assessment.Stat Methods Med Res2015;
24(1): 27–67.
3.
KesslerLGBarnhartHXBucklerAJ, et al.
The emerging science of quantitative imaging biomarkers terminology and definitions for scientific studies and regulatory submissions.Stat Methods Med Res2015;
24: 9–26.
4.
ObuchowskiNAReevesAPHuangEP, et al.
Quantitative imaging biomarkers: A review of statistical methods for computer algorithm comparisons.Stat Methods Med Res2015;
24: 68–106.
5.
ObuchowskiNABarnhartHXBucklerAJ, et al.
Statistical issues in the comparison of quantitative imaging biomarker algorithms using pulmonary nodule volume as an example.Stat Methods Med Res2015;
24: 107–140.
6.
CohenPWestSGAikenLS.Applied multiple regression/correlation analysis for the behavioral sciences.
London:
Psychology Press, 2014.
7.
MarascuiloLALevinJR.Multivariate statistics in the social sciences: A researcher’s guide.
Belmont:
Wadsworth Publishing Company, 1983.
8.
NeterJKutnerMHNachtsheimCJ, et al.Applied linear statistical models.
New York:
Irwin, 1996.
9.
PedhazurEJKerlingerFN.Multiple regression in behavioral research: Explanation and prediction.
New York:
Holt, Rinehart, and Winston, 1982.
StoneEFHollenbeckJR.Clarifying some controversial issues surrounding statistical procedures for detecting moderator variables: Empirical evidence and related matters.J Appl Psychol1989;
74: 3.
12.
LinSMDuPHuberW, et al.
Model-based variance-stabilizing transformation for Illumina microarray data.Nucleic Acids Res2008;
36: e11.
13.
CurtissJH.On transformations used in the analysis of variance. Ann Math Stat1943;
14: 107–122.
14.
HuberWVon HeydebreckASültmannH, et al.
Variance stabilization applied to microarray data calibration and to the quantification of differential expression. Bioinformatics2002;
18: S96–S104.
15.
HuberWvon HeydebreckASültmannH, et al.
Parameter estimation for the calibration and variance stabilization of microarray data. Stat Appl Genet Mol Biol2003;
2.
16.
RockeDMDurbinB.Approximate variance-stabilizing transformations for gene-expression microarray data.Bioinformatics2003;
19: 966–972.
17.
DurbinBPHardinJSHawkinsDM, et al.
A variance-stabilizing transformation for gene-expression microarray data. Bioinformatics2002;
18: S105–S110.
18.
BartlettJFrostC.Reliability, repeatability and reproducibility: Analysis of measurement errors in continuous variables.Ultrasound Obstet Gynecol2008;
31: 466–475.
19.
BenoitK.Linear regression models with logarithmic transformations. London School of Economics2011;
22: 23–36.
20.
GongQLiQGavrielidesMA, et al. Data transformations for variance stabilization in the statistical assessment of quantitative imaging biomarkers. In: Medical imaging 2019: Image perception, observer performance, and technology assessment, San Diego, CA, 16–21 February 2019, p.109520Z. Bellingham: International Society for Optics and Photonics.
21.
SakiaRM.The Box-Cox transformation technique – A review. J Roy Stat Soc D1992;
41: 169–178.
22.
TherassePArbuckSGEisenhauerEA, et al.
New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada.J Natl Cancer Inst2000;
92: 205–216.
23.
GavrielidesMAKinnardLMMyersKJ, et al.
Non-calcified lung nodules: Volumetric assessment with thoracic CT. Radiology2009;
251: 26–37.
24.
GavrielidesMALiQZengR, et al.
Volumetric analysis of non-calcified lung nodules with thoracic CT: An updated review of related work over the last 5 years. J Radiol Radiat Ther2014;
2: 1040.
25.
GavrielidesMALiQZengR, et al. Detectable change of lung nodule volume with CT in a phantom study with high and low signal to background contrast. In: SPIE medical imaging, San Diego, CA, 25 March 2016, p.8. Bellingham: SPIE.
26.
GavrielidesMALiQZengR, et al.
Discrimination of pulmonary nodule volume change for low-and high-contrast tasks in a phantom CT study with low-dose protocols. Acad Radiol2019:
26(7): 937–948.
27.
LiQGavrielidesMASahinerB, et al.
Statistical analysis of lung nodule volume measurements with CT in a large-scale phantom study.Med Phys2015;
42: 3932–3947.
28.
MakitaloMFoiA.Optimal inversion of the Anscombe transformation in low-count Poisson image denoising.IEEE Trans Image Process2011;
20: 99–109.
29.
GoodsittMMChanH-PWayTW, et al.
Accuracy of the CT numbers of simulated lung nodules imaged with multi-detector CT scanners.Med Phys2006;
33: 3006–3017.
30.
GurcanMNSahinerBPetrickN, et al.
Lung nodule detection on thoracic computed tomography images: Preliminary evaluation of a computer-aided diagnosis system.Med Phys2002;
29: 2552–2558.
31.
ShiraishiJKatsuragawaSIkezoeJ, et al.
Development of a digital image database for chest radiographs with and without a lung nodule: Receiver operating characteristic analysis of radiologists’ detection of pulmonary nodules.Am J Roentgenol2000;
174: 71–74.
32.
BucklerAJDanagoulianJJohnsonK, et al.
Inter-method performance study of tumor volumetry assessment on computed tomography test-retest data.Acad Radiol2015;
22: 1393–1408.
33.
ZhaoBJamesLPMoskowitzCS, et al.
Evaluating variability in tumor measurements from same-day repeat CT scans of patients with non-small cell lung cancer.Radiology2009;
252: 263–272.
34.
CarpenterJBithellJ.Bootstrap confidence intervals: When, which, what? A practical guide for medical statisticians. Stat Med2000;
19: 1141–1164.
35.
StephensMA.EDF statistics for goodness of fit and some comparisons. J Am Stat Assoc1974;
69: 730–737.
36.
AndersonTWDarlingDA.A test of goodness of fit. J Am Stat Assoc1954;
49: 765–769.
37.
LiQGavrielidesMAZengR, et al. Factors affecting uncertainty in lung nodule volume estimation with CT: Comparisons of findings from two estimation methods in a phantom study. In: SPIE medical imaging 2015: Computer-aided diagnosis (eds Hadjiiski L and Tourassi GD), Orlando, FL, 2015. Bellingham: SPIE.
