Contrast reagent leakage correction in dynamic susceptibility contrast magnetic resonance imaging
Disclosed are methods and systems for calculating a contrast reagent (CR) extravasation rate constant and generating a contrast reagent leakage corrected relative cerebral blood volume (rCBV) image map of a brain region from dynamic susceptibility contrast (DSC) magnetic resonance imaging (MRI) time-course image data based on pharmacokinetic first principles. In one example approach, a computerized method may include performing a linearization transform of a DSC MRI time-course equation which accounts for an intravascular contribution and an extravasating component, and calculating CR leakage from a slope of a linear portion of the transformed data.
1. A computerized method for contrast reagent leakage correction of dynamic susceptibility contrast (DSC) magnetic resonance imaging (MRI) time-course image data of a brain region, the method comprising:
from the DSC MRI time-course image data, computing a pixel time-course of an effective transverse relaxation rate constant change (ΔR 2 *(t)) and an estimate of a time-course of an effective transverse relaxation rate constant change for relative blood ( ΔR 2 * (t));
computing a contrast reagent extravasating rate constant (K L ) from a slope of a linear portion of a mapping of ΔR 2 *(t)/ ΔR 2 * (t) versus ∫ 0 t ΔR 2 *(t′)dt′/ΔR 2 *(t);
generating a relative cerebral blood volume (rCBV) image map;
generating a contrast reagent leakage corrected relative cerebral blood volume (rCBV LC ) image map based on the contrast reagent extravasating rate constant (K L ),
wherein the contrast reagent leakage corrected rCBV LC image map is generated using the equation:
rCBV LC =rCBV un +K L ∫ 0 T dt′∫ 0 t′ Δ R 2 * ( t ) dt
where rCBV LC is the leakage-corrected rCBV, rCBV un is uncorrected rCBV, K L is the contrast reagent extravasating rate constant, and ΔR 2 * (t) is the effective transverse relaxation rate constant change for relative blood.
2. The method of claim 1 , wherein the contrast reagent comprises a gadolinium-based contrast reagent.
3. The method of claim 1 , wherein the linear portion of the mapping corresponds to DSC MRI time-course image data acquired following a transient period of contrast reagent first pass.
4. The method of claim 1 , wherein ΔR 2 *(t) is computed based on a baseline signal intensity and an imaging echo time of a pulse sequence used to acquire the DSC MRI data.
5. The method of claim 1 , wherein ΔR 2 * (t) is computed based on a combination of signal changes from non-leaking pixels within the DSC MRI time-course image data.
6. The method of claim 1 , wherein the DSC MRI time-course image data comprises dynamic T 2 *-weighted images of the brain region acquired using a gradient-echo echo planar imaging pulse sequence.
7. The method of claim 1 , wherein ΔR 2 *(t) is computed according to the equation:
Δ
R
2
*
(
t
)
=
-
ln
(
S
(
t
)
S
0
)
TE
where S(t) is the DSC MRI pixel image at time t, S 0 is a baseline signal intensity, and TE is an imaging echo time of a pulse sequence used to acquire the DSC MRI data.
8. The method of claim 1 , wherein the linear portion of the mapping corresponds to DSC MRI time-course image data acquired from approximately 40 seconds after injection of the contrast reagent to approximately 70 seconds after injection of the contrast reagent.
9. A method for producing an image of a patient's brain with a magnetic resonance imaging (MRI) system, comprising:
acquiring dynamic susceptibility contrast (DSC) MRI time-course image data for a region of interest of the brain prior to an injection of a contrast reagent to obtain a baseline signal intensity;
acquiring DSC MRI time-course image data for the region of interest following an injection of a contrast reagent into a blood vessel which induces a differential between the susceptibilities of the blood vessel and tissue in the region of interest;
from the DSC MRI time-course image data, calculating a pixel time-course of an effective transverse relaxation rate constant change (ΔR 2 *(t)) based on the baseline signal intensity and an imaging echo time of a pulse sequence used to acquire the DSC MRI data;
calculating an estimate of a time-course of an effective transverse relaxation rate constant change for relative blood ( ΔR 2 * (t)) based on a combination of signal changes from non-leaking pixels within the DSC MRI time-course image data;
calculating a contrast reagent extravasating rate constant (K L ) from a slope of a linear portion of a mapping of ΔR 2 *(t)/ ΔR 2 * (t) versus ∫ 0 t ΔR 2 * (t′)dt′/ ΔR 2 * (t), where the linear portion of the mapping corresponds to DSC MRI time-course image data acquire following a transient period of contrast reagent first pass;
generating an image map of relative cerebral blood volume (rCBV) in the region of interest based on an integration of ΔR 2 *(t) over the DSC MRI time-course image data for each pixel in the region of interest;
generating a corrected rCBV image map to correct the rCBV image map for leakage of the contrast reagent out of the patient's vasculature based on the contrast reagent extravasating rate constant (K L ); and
outputting the corrected rCBV image map to a display device.
10. The method of claim 9 , wherein the contrast reagent comprises a gadolinium-based contrast reagent.
11. The method of claim 9 , wherein acquiring DSC MRI time-course image data for the region of interest comprises acquiring dynamic T 2 *-weighted images of the region of interest using a gradient-echo echo planar imaging pulse sequence.
12. The method of claim 9 , wherein ΔR 2 *(t) is calculated according to the equation:
Δ
R
2
*
(
t
)
=
-
ln
(
S
(
t
)
S
0
)
TE
where S(t) is the DSC MRI pixel image at time t, S 0 is the baseline signal intensity, and TE is the imaging echo time.
13. The method of claim 9 , wherein the linear portion of the mapping corresponds to DSC MRI time-course image data acquired from approximately 40 seconds after injection of the contrast reagent to approximately 70 seconds after injection of the contrast reagent.
14. The method of claim 9 , wherein the corrected rCBV image map is generated according to the equation:
rCBV LC =rCBV un +K L ∫ 0 T dt′∫ 0 t′ Δ R 2 * ( t ) dt
where rCBV LC is the leakage corrected rCBV, rCBV un is uncorrected rCBV, K L is the contrast reagent extravasating rate constant, and ΔR 2 * (t) is the effective transverse relaxation rate constant change for relative blood.
15. The method of claim 9 , further comprising motion correcting the DSC MRI time-course image data.
16. The method or claim 9 , further comprising performing a linear regression on the linear portion of the mapping to obtain the contrast reagent extravasating rate constant.
17. The method of claim 9 , further comprising identifying an intravasation period from the mapping in response to a deviation from linearity in the mapping following the transient period of contrast reagent first pass.