DOI: 10.1148/radiol.2462070655
(Radiology 2008;246:642-643.)
© RSNA, 2008
Are Measurements from Two Commercial Software Packages Interchangeable? Possibly, If Like Is Compared with Like
David L. Buckley, PhD
Imaging Science and Biomedical Engineering, University of Manchester, Stopford Building, Oxford Road, Manchester M13 9PT, England
e-mail: david.buckley{at}manchester.ac.uk
Editor:
In the March 2007 issue of Radiology, Dr Goh and colleagues (1) compared contrast material–enhanced computed tomographic (CT) measurements of tumor vascularity obtained with two commercial software packages. The two packages produced differing results and it was concluded that the techniques were not directly interchangeable. While considering the differences between the two indexes of tissue permeability, the authors assess the models as follows: "[T]here are conceptual and mathematic differences between them that may contribute to disagreement" (1). What they have actually compared are estimates of permeability–surface area product (PS) (using the distributed parameter model) and extraction-flow product (EF) (using Patlak analysis). Thus the packages should not be expected to produce interchangeable results; the authors have not compared like with like.
It is not clear from the article precisely why disparate parameters were compared at the outset. These are neither the parameters nor the pairing of models described by Miles and Griffiths (2) as "broadly equivalent." Moreover, the use of the word "perfusion" in the title, abstract, and Advance in Knowledge of the article by Dr Goh and colleagues (1) adds to the confusion, since perfusion estimates per se are not compared at any stage in this study.
Comparing commercial software packages is challenging, and the authors should be applauded for undertaking such a study. Nevertheless, it is possible to compare like with like (eg, see reference 3) and, with modification, the outputs of these two packages. If flow, F, is much greater than PS, then the Patlak parameter EF is approximately equal to PS (4), although this is rarely known a priori and does not appear to be the case in colorectal tumors (5). Alternatively, the distributed parameter model can provide its own estimates of EF through combination of estimates of F and PS; EF = F · {1 – exp[–PS/F(1 – Hct)]}, where Hct is hematocrit (4).
 |
References
|
|---|
- Goh V, Halligan S, Bartram CI. Quantitative tumor perfusion assessment with multidetector CT: are measurements from two commercial software packages interchangeable? Radiology 2007;242:777–782. [Abstract/Free Full Text]
- Miles KA, Griffiths MR. Perfusion CT: a worthwhile enhancement? Br J Radiol 2003;76:220–231. [Free Full Text]
- Buckley DL. Uncertainty in the analysis of tracer kinetics using dynamic contrast-enhanced T1-weighted MRI. Magn Reson Med 2002;47:601–606. [CrossRef][Medline]
- Tofts PS, Brix G, Buckley DL, et al. Estimating kinetic parameters from dynamic contrast-enhanced T1-weighted MRI of a diffusable tracer: standardized quantities and symbols. J Magn Reson Imaging 1999;10:223–232. [CrossRef][Medline]
- Sahani DV, Kalva SP, Hamberg LM, et al. Assessing tumor perfusion and treatment response in rectal cancer with multisection CT: initial observations. Radiology 2005;234:785–792.[Abstract/Free Full Text]
Response
Vicky Goh, MA, MRCP, FRCR * and
Steve Halligan, MD, FRCP, FRCR 
* Paul Strickland Scanner Centre, Mount Vernon Hospital, Rickmansworth Road, Northwood, Middlesex HA6 2RN, England
e-mail: vicky.goh{at}stricklandscanner.org.uk
Department of Academic Radiology, University College London, University College Hospital, London, England
Dr Buckley has raised a valid mathematical point: PS (using the distributed parameter model) and EF (using Patlak analysis) are not the same measurement. Dr Buckley states that it is not clear "why disparate parameters were compared at the outset." We did so because we wanted to adopt the stance of the end-user—that is, radiologists who use commercially available software in their day-to-day clinical practice. Taking this perspective, the disparity to which Dr Buckley refers is obscured by manufacturers' claims that each measures "perfusion parameters" including blood flow, blood volume, and PS.
We agree that when flow is much greater than PS, then Patlak EF may reflect PS, and this assumption underpins commercial marketing as such. Dr Buckley suggests that, a priori, this is an unknown and thus cannot be applied for colorectal cancer. However there are available data, from both ourselves (1) and others (2), that provide estimates of flow in colorectal cancer and on which assumptions can be based. Dr Buckley has suggested a sensible method with which to adjust for different software packages, should direct comparison be necessary in clinical practice. Whether radiologists will adopt this in their day-to-day practice is questionable but might change should manufacturers adopt Dr Buckley's suggestion.
We chose not to compare flow directly in our study because acquisition parameters might favor distributed parameter analysis. Until terminology is standardized, a development that we would applaud, usage of terms such as "perfusion CT" instead of "dynamic contrast-enhanced CT" to refer to both first-pass and delayed phase acquisitions or "perfusion measurements" to refer to vascular measurements in general (including blood volume or PS) rather than the strictly correct blood flow per unit volume of tissue will inevitably irritate purists (3).
|
References
|
|---|
- Goh V, Halligan S, Hugill JA, Bartram CI. Quantitative assessment of tissue perfusion using MDCT: comparison of colorectal cancer and skeletal muscle measurement reproducibility. AJR Am J Roentgenol 2006;187:164–169. [Abstract/Free Full Text]
- Sahani DV, Kalva SP, Hamberg LM, et al. Assessing tumor perfusion and treatment response in rectal cancer with multisection CT: initial observations. Radiology 2005;234:785–792. [Abstract/Free Full Text]
- Miles KA, Griffiths MR. Perfusion CT: a worthwhile enhancement? Br J Radiol 2003;76:220–231.[Free Full Text]
TOP
References
References 
