Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Table 1.

Sequence parameters.

More »

Table 1 Expand

Table 2.

Fit algorithms.

More »

Table 2 Expand

Fig 1.

Example parameter maps from a pancreatic cancer patient.

Axial parameter maps of the fit parameters of six IVIM model fit algorithms in a 60-year-old female with pancreatic adenocarcinoma in the pancreas tail. ROIs containing pancreatic tumour are shown. The CE T1W GE is added as a reference. Note that for the Bayesian approaches, not all voxels were fitted, as including more voxels will influence the prior. This patient had D, f and D* between 1.1–1.5×10−3 mm2/s, 1.1–2.1% and 116–989×10−3 mm2/s respectively. The yellow highlights the high D* values fitted in IVIM-Bayesian-log, compared to the other fit algorithms.

More »

Fig 1 Expand

Fig 2.

Example parameter maps from a pancreatic cancer patient.

Axial parameter maps of the fit parameters of six IVIM model fit algorithms in a 61-year-old female with pancreatic adenocarcinoma in the pancreas corpus. ROIs containing pancreatic tumour are shown. The CE T1W GE is added as a reference. Note that for the Bayesian approaches, not all voxels were fitted, as including more voxels will influence the prior. This patient had D, f and D* between 1.3–1.5×10−3 mm2/s, 2.1–6.6% and 43–98×10−3 mm2/s respectively. The green arrow highlights the higher f found in IVIM-Bayesian compared to the other algorithms.

More »

Fig 2 Expand

Fig 3.

Example parameter maps from a pancreatic cancer patient.

Axial parameter maps of the fit parameters of six IVIM model fit algorithms in a 71-year-old male with pancreatic adenocarcinoma in the pancreas tail. ROIs containing pancreatic tumour are shown. The CE T1W GE is added as a reference. Note that for the Bayesian approaches, not all voxels were fitted, as including more voxels will influence the prior. This patient had D, f and D* between 1.0–1.3×10−3 mm2/s, 2.0–14.9% and 15–70×10−3 mm2/s respectively. The green arrow highlights the higher f found in IVIM-Bayesian compared to the other algorithms. The blue arrows highlight the lower f in IVIM-Bayesian-lin and IVIM-fix.

More »

Fig 3 Expand

Fig 4.

Correlations between fit parameters.

Correlation between fit parameters D&f (left column), D&D* (middle column) and f&D* (right column). For the parameter pairs with significant correlation according to Spearman’s rank correlation, a linear regression line is plotted to the data from pancreatic cancer. The correlations were only tested using pancreatic cancer tissue (dark red dots). Note that the D*-axis of IVIM-Bayesian-log is stretched to fit all data and hence highlighted in red.

More »

Fig 4 Expand

Table 3.

Uniqueness.

More »

Table 3 Expand

Table 4.

Precision of parameters.

More »

Table 4 Expand

Table 5.

Contrast.

More »

Table 5 Expand

Fig 5.

wCV vs contrast.

Plots of the inter-session wCV as a function of contrast between tumour and pancreatic tissue for the diffusion-related parameter (blue) and other fit parameters (red, green). Bottom right graph has zoomed to low wCV and contrast to illustrate the trade-off for D.

More »

Fig 5 Expand