The Annals of Applied Statistics

Applying a spatiotemporal model for longitudinal cardiac imaging data

Brandon George, Thomas Denney, Jr., Himanshu Gupta, Louis Dell’Italia, and Inmaculada Aban

Full-text: Open access

Abstract

Longitudinal imaging studies have both spatial and temporal correlation among the multiple outcome measurements from a subject. Statistical methods of analysis must properly account for this autocorrelation. In this work we discuss how a linear model with a separable parametric correlation structure could be used to analyze data from such a study. The goal of this paper is to provide an easily understood description of how such a model works and discuss how it can be applied to real data. Model assumptions are discussed and the process of selecting a working correlation structure is thoroughly discussed. The steps necessitating collaboration between statistical and scientific investigators have been highlighted, as have considerations for missing data or uneven follow-up.

The results from a completed longitudinal cardiac imaging study were considered for illustration purposes. The data comes from a clinical trial for medical therapy for patients with mitral regurgitation, with repeated measurements taken at sixteen locations from the left ventricle to measure disease progression. The spatiotemporal correlation model was compared to previously used summary measures to demonstrate improved power as well as increased flexibility in the use of time- and space-varying predictors.

Article information

Source
Ann. Appl. Stat., Volume 10, Number 1 (2016), 527-548.

Dates
Received: October 2015
First available in Project Euclid: 25 March 2016

Permanent link to this document
https://projecteuclid.org/euclid.aoas/1458909926

Digital Object Identifier
doi:10.1214/16-AOAS911

Mathematical Reviews number (MathSciNet)
MR3480506

Keywords
Spatiotemporal imaging correlation separable summary measures

Citation

George, Brandon; Denney, Jr., Thomas; Gupta, Himanshu; Dell’Italia, Louis; Aban, Inmaculada. Applying a spatiotemporal model for longitudinal cardiac imaging data. Ann. Appl. Stat. 10 (2016), no. 1, 527--548. doi:10.1214/16-AOAS911. https://projecteuclid.org/euclid.aoas/1458909926


Export citation

References

  • Ahmed, M. I., Gladden, J. D., Litovsky, S. H., Lloyd, S. G., Gupta, H., Inusah, S., Denny, T. Jr., Powell, P., McGiffin, D. C. and Dell’Italia, L. J. (2010). Increased oxidative stress and cardiomyocyte myofibrillar degeneration in patients with chronic isolated mitral regurgitation and ejection fraction $>60$%. J. Am. Coll. Cardiol. 55 671–679.
  • Ahmed, M. I., Aban, I., Lloyd, S. G., Gupta, H., Howard, G., Inusah, S., Peri, K., Robinson, J., Smith, P., McGiffin, D. C., Schiros, C. G., Denney, T. S. and Dell’Italia, L. J. (2012). A randomized controlled phase IIb trial of beta-1-receptor blockade for chronic degenerative mitral regurgitation. J. Am. Coll. Cardiol. 60 833–838.
  • Allison, P. D. (2012). Handling missing data by maximum likelihood. SAS Global Forum 2012, Paper 312-2012, Available at http://statisticalhorizons.com/wp-content/uploads/MissingDataByML.pdf.
  • Bates, D., Maechler, M., Bolker, B. and Walker, S. (2014). Package ‘lme4’ [Software]. Available at http://lme4.r-forge.r-project.org/.
  • Beyar, R., Weiss, J. L., Shapiro, E. P., Graves, W. L., Rogers, W. J. and Weisfeldt, M. L. (1993). Small apex-to-base heterogeneity in radius-to-thickness ratio by three-dimensional magnetic resonance imaging. Am. J. Physiol. 264 H133–H140.
  • Bowman, F. D. and Waller, L. A. (2004). Modelling of cardiac imaging data with spatial correlation. Stat. Med. 23 965–985.
  • Butler, D., Cullis, B. R., Gilmour, A. R. and Gogel, B. J. (2007). ASReml-R reference manual (Release 2.00) [Software]. Available at http://www.vsni.co.uk/software/asreml.
  • Cerqueria, M. D., Weissman, N. J., Dilsizian, V., Jacobs, A. K., Kaul, S., Laskey, W. K., Pennell, D. J., Rumberger, J. A., Ryan, T. and Verani, M. S. (2002). Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: A statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation 105 539–542.
  • Everitt, B. S. (1995). In the analysis of repeated measures: A practical review with examples. Statistician 44 113–135.
  • Fitzmaurice, G. M., Laird, N. M. and Ware, J. H. (2004). Applied Longitudinal Analysis. Wiley, Hoboken, NJ.
  • Frison, L. and Pocock, S. J. (1992). Repeated measures in clinical trials: Analysis using mean summary statistics and its implications for design. Stat. Med. 11 1685–1704.
  • George, B. (2014). A spatiotemporal model for repeated imaging data. Ph.D. dissertation, Univ. Alabama at Birmingham, Birmingham, AL.
  • George, B. and Aban, I. (2015). Selecting a separable parametric spatiotemporal covariance structure for longitudinal imaging data. Stat. Med. 34 145–161.
  • George, B., Denney, Jr., T., Gupta, H., Dell’Italia, L. and Aban, I. (2016). Supplement to “Applying a spatiotemporal model for longitudinal cardiac imaging data.” DOI:10.1214/16-AOAS911SUPP.
  • Gibbons, R., Hedecker, D., Waternaux, C. and Davis, J. M. (1988). Random regression models: A comprehensive approach to the analysis of longitudinal psychiatric data. Psychopharmacol. Bull. 24 438–443.
  • Kenward, M. G. and Roger, J. H. (1997). Small sample inference for fixed effects from restricted maximum likelihood. Biometrics 53 983–997.
  • Littell, R. C., Pendergast, J. and Natarajan, R. (2000). Modelling covariance structure in the analysis of repeated measures data. Stat. Med. 19 1793–1819.
  • Mann, D. L., Kend, R. L., Parsons, B. and Cooper, G. (1992). Adrenergic effects on the biology of the adult mammalian cardiocyte. Circulation 85 790–804.
  • Matthews, J. N. S. (1990). A refinement to the analysis of serial data using summary measures. Stat. Med. 12 27–37.
  • Nagatsu, M., Zile, M. R., Tsutsui, H., Schmid, P. S., DeFreyte, D., Cooper, G. and Carabello, B. A. (1994). Native $\beta$-adrenergic support for left ventricular dysfunction in experimental mitral regurgitation normalizes indexes of pump and contractile function. Circulation 89 818–826.
  • SAS Institute Inc. (2014). SAS/STAT 13.2 User’s Guide. SAS Institute Inc.: Cary, NC, 2014.
  • Schaalje, B., McBride, J. B. and Fellingham, G. W. (2002). Adequacy of approximations to distributions of test statistics in complex mixed linear models. J. Agric. Biol. Environ. Stat. 7 512–524.
  • Schiros, C. G., Dell’Italia, L. J., Gladden, J. D., Clark, D., Aban, I., Gupta, H., Lloys, S. G., McGiffin, D. C., Perry, G., Denney, T. S. and Ahmed, M. I. (2012). Magnetic resonance imaging with 3-dimensional analysis of left ventricular remodeling in isolated mitral regurgitation: Implications beyond dimensions. Circulation 125 2334–2342.
  • Seals, S., Katholi, C. R., Zhang, J. and Aban, I. B. (2014). Evaluating the use of spatial covariance structures in the analysis of cardiovascular imaging data. Comm. Statist. Simulation Comput. DOI:10.1080/03610918.2014.925922
  • Simpson, S. L., Edwards, L. J., Muller, K. E. and Styner, M. A. (2014a). Kronecker product exponent AR(1) correlation structures for multivariate repeated measures. PLoS ONE 9 e88864.
  • Simpson, S. L., Edwards, L. J., Styner, M. A. and Muller, K. E. (2014b). Separability tests for high-dimensional, low-sample size multivariate repeated measures data. J. Appl. Stat. 41 2450–2461.
  • Standaert, D. (2014). Image analysis in clinical practice: Measuring progress. Presented at: Advanced imaging analysis. In Proceedings of the AMC 21 Research Strategic Planning Committee Imaging Symposium; 2014 Apr 1. Birmingham, AL.
  • Tsutsui, H., Spinale, F. G., Nagatsu, M., Schmid, P. S., Ishihara, K., DeFreyte, G., Cooper, G. and Carabello, B. A. (1994). Effects of chronic $\beta$-adrenergic blockade on the left ventricular and cardiocyte abnormalities of chronic canine mitral regurgitation. J. Clin. Invest. 93 2639–2648.
  • Vallejo, G., Fernández, M. P., Livacic-Rojas, P. E. and Tuero-Herrero, E. (2011). Selecting the best unbalanced repeated measures model. Behav. Res. Methods 43 18–36.
  • Waller, L. A. and Gotway, C. A. (2004). Applied Spatial Statistics for Public Health Data. Wiley, Hoboken, NJ.
  • Young, A. A., Orr, R., Smaill, B. H. and Dell’Italia, L. J. (1996). Three-dimensional changes in left and right ventricular geometry in chronic mitral regurgitation. Am. J. Physiol. Heart Circ. Physiol. 271 H2689–H2700.

Supplemental materials

  • Supplementary tables and figure. We provide QQ plots of the model residuals justifying the log transformation, and tables of the regression parameters for the primary spatiotemporal model and inference about a treatment-by-time interaction when the post-surgery observations were excluded from the analysis.