Journal
EUROPEAN RADIOLOGY
Volume 28, Issue 4, Pages 1485-1494Publisher
SPRINGER
DOI: 10.1007/s00330-017-5094-3
Keywords
Rectal cancer; Diffusion-weighted imaging; Diffusion kurtosis imaging; Prognostic factors; Biomarker
Funding
- Science and Technology Project of Shanxi Province [20150313007-5]
Ask authors/readers for more resources
To investigate potential relationships between diffusion kurtosis imaging (DKI)-derived parameters using whole-tumour volume histogram analysis and clinicopathological prognostic factors in patients with rectal adenocarcinoma. 79 consecutive patients who underwent MRI examination with rectal adenocarcinoma were retrospectively evaluated. Parameters D, K and conventional ADC were measured using whole-tumour volume histogram analysis. Student's t-test or Mann-Whitney U-test, receiver operating characteristic curves and Spearman's correlation were used for statistical analysis. Almost all the percentile metrics of K were correlated positively with nodal involvement, higher histological grades, the presence of lymphangiovascular invasion (LVI) and circumferential margin (CRM) (p < 0.05), with the exception of between K-10th, K-90th and histological grades. In contrast, significant negative correlations were observed between 25th, 50th percentiles and mean values of ADC and D, as well as ADC(10th), with tumour T stages (p < 0.05). Meanwhile, lower 75th and 90th percentiles of ADC and D values were also correlated inversely with nodal involvement (p < 0.05). K-mean showed a relatively higher area under the curve (AUC) and higher specificity than other percentiles for differentiation of lesions with nodal involvement. DKI metrics with whole-tumour volume histogram analysis, especially K parameters, were associated with important prognostic factors of rectal cancer. aEuro cent K correlated positively with some important prognostic factors of rectal cancer. aEuro cent K (mean) showed higher AUC and specificity for differentiation of nodal involvement. aEuro cent DKI metrics with whole-tumour volume histogram analysis depicted tumour heterogeneity.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available