Intravoxel incoherent motion diffusion‐weighted MR imaging of breast cancer at 3.0 tesla: Comparison of different curve‐fitting methods. Issue 2 (19th November 2014)

• Record Type: Journal Article
• Authors: Suo, Shiteng; Lin, Naier; Wang, He; Zhang, Liangbin; Wang, Rui; Zhang, Su; Hua, Jia; Xu, Jianrong
• Is Part Of: Journal of magnetic resonance imaging. Volume 42:Issue 2(2015)
• Page Start: 362
• Page End: 370
• Publication Date: 2014-11-19
• Subjects: Magnetic resonance imaging -- Periodicals; 616
• DOI: 10.1002/jmri.24799
• Languages: English
• ISSNs: 1053-1807

Abstract

<abstract abstract-type="main"> <title> <x xml:space="preserve">Abstract</x> </title> <sec id="jmri24799-sec-1001" sec-type="section"> <title>Background</title> <p>To compare three different curve‐fitting methods for intravoxel incoherent motion (IVIM) analysis in breast cancer.</p> </sec> <sec id="jmri24799-sec-1002" sec-type="section"> <title>Methods</title> <p>Diffusion‐weighted imaging was acquired in 30 patients with breast cancer using seven <italic>b</italic>‐values (0–800 s/mm<sup>2</sup>). Three curve‐fitting methods were used for biexponential IVIM analysis: a. Direct estimation of <italic>D</italic> (diffusion coefficient), <italic>D*</italic> (pseudodiffusion coefficient) and <italic>f</italic> (perfusion fraction) (Method 1), b. Estimation of <italic>D</italic> first and then <italic>D*</italic> and <italic>f</italic> (Method 2), c. Estimation of <italic>D</italic> and <italic>f</italic> first and then <italic>D*</italic> (Method 3). Goodness‐of‐fit, parameter precision (coefficient of variance [CV]), parameter difference and correlation with relative enhancement ratio (RER) and initial area under the curve (IAUC) from dynamic contrast‐enhanced (DCE) MRI of the three methods were determined and compared.</p> </sec> <sec id="jmri24799-sec-1003" sec-type="section"> <title>Results</title> <p>Among the three biexponential methods, Method 1 best described most of the pixels (63.20% based on <italic>R<sup>2</sup></italic>; 44.52% based on Akaike Information Criteria). The CV of <italic>D</italic> calculated from Method 2/3 (14.95%/13.90%), the CV of <italic>D*</italic> from Method 2 (77.04%) and the CV of <italic>f</italic> from Method 3 (80.87%) were the lowest among the three methods. Significant difference was observed for each IVIM‐derived parameter calculated from all the three methods (<italic>P</italic> = 0.000–0.005). Only the perfusion‐related <italic>f</italic> value calculated from Method 2 was correlated with RER (<italic>r</italic> = 0.548; <italic>P</italic> = 0.002) or IAUC (<italic>r</italic> = 0.561; <italic>P</italic> = 0.001).</p> </sec> <sec id="jmri24799-sec-1004" sec-type="section"> <title>Conclusion</title> <p>IVIM‐derived parameters differ depending on the calculation methods. The two‐step fitting method with <italic>D</italic> value estimation first was correlated with DCE MRI perfusion. J. Magn. Reson. Imaging 2015;42:362–370.</p> </sec> </abstract>