Bland-Altman plot (Tukey mean-difference plot) is a simple method for analyzing the agreement between two different analysis methods or interobserver variability. Comparison can even be done pixel-by-pixel for parametric images.
Bland-Altman plot is made by plotting the average of the two values to be compared to the x axis, and the difference to the y axis.
Bland-Altman plots are also simple to interpret:
- If there is (significant) bias, is it clinically important?
- If there is a trend, then in what direction it would affect your results?
- Is variance increasing with increasing average?
Standard deviation (SD) of the differences between the two methods can also be calculated, and further used to describe the limits of agreement, as the mean ± 1.96×SD.
Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; i: 307-310.
de Vet HCW, Terwee CB, Mokkink LB, Knol DL (eds.): Measurement in Medicine: A Practical Guide. Cambridge University Press, 2011. ISBN: 9780521118200.
Eksborg S. Evaluation of method-comparison data. Clin Chem. 1981; 27(7): 1311-1312.
Gerke O, Vilstrup MH, Segtnan EA, Halekoh U, Høilund-Carsen PF. How to assess intra- and inter-observer agreement with quantitative PET using variance component analysis: a proposal for standardisation. BMC Med Imaging 2016; 16:54.
Nesterov SV, Han C, Mäki M, Kajander S, Naum AG, Helenius H, Lisinen I, Ukkonen H, Pietilä M, Joutsiniemi E, Knuuti J. Myocardial perfusion quantitation with 15O-labelled water PET: high reproducibility of the new cardiac analysis software (Carimas). Eur J Nucl Med Mol Imaging. 2009; 36(10): 1594-1602.
Created at: 2014-05-23
Updated at: 2017-01-30
Written by: Vesa Oikonen