Back to Search Start Over

Discrimination of Malignant versus Benign Mediastinal Lymph Nodes Using Diffusion MRI with an IVIM Model.

Authors :
Qi, Li-Ping
Yan, Wan-Pu
Chen, Ke-Neng
Zhong, Zheng
Li, Xiao-Ting
Cai, Kejia
Sun, Ying-Shi
Zhou, Xiaohong Joe
Source :
European Radiology. Mar2018, Vol. 28 Issue 3, p1301-1309. 9p. 3 Charts, 3 Graphs.
Publication Year :
2018

Abstract

<bold>Objectives: </bold>To investigate the value of an intravoxel incoherent motion (IVIM) diffusion model for discriminating malignant versus benign mediastinal lymph nodes (MLN).<bold>Methods: </bold>Thirty-five subjects with enlarged MLN were scanned at 1.5 Tesla. Diffusion-weighted imaging was performed with eight b-values. IVIM parameters D, D*, and f, as well as apparent diffusion coefficient (ADC) from a mono-exponential model were obtained. 91 nodes (49 malignant and 42 benign) were analysed with pathologic (n=90) or radiologic (n=1) confirmations. Receiver operating characteristic (ROC) analysis was used to evaluate the diagnostic performance.<bold>Results: </bold>The mean values of D, ADC, and f for the malignant group were significantly lower than those for the benign group (p<0.001), while D* showed no significant difference (p=0.281). In the ROC analysis, the combination of D and f produced the largest area under the curve (0.953) compared to ADC or other individual IVIM parameters, leading to the best specificity (92.9%) and diagnostic accuracy (90.1%).<bold>Conclusion: </bold>This study demonstrates that the combination of IVIM parameters can improve differentiation between malignant and benign MLN as compared to using ADC alone.<bold>Key Points: </bold>• Diffusion MRI is useful for non-invasively discriminating malignant versus benign lymph nodes. • A mono-exponential model is not adequate to characterise diffusion process in lymph nodes. • IVIM model is advantageous over mono-exponential model for assessing lymph node malignancy. • Combination of IVIM parameters improves differentiation of malignant versus benign lymph nodes. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
09387994
Volume :
28
Issue :
3
Database :
Academic Search Index
Journal :
European Radiology
Publication Type :
Academic Journal
Accession number :
127990348
Full Text :
https://doi.org/10.1007/s00330-017-5049-8