Edema invariant tractography
AssigneeThe Trustees of the University of Pennsylvania
InventorsRagini Verma, Luke Macyszyn, Emmanuel Caruyer, Jeremy Lecoeur, Mark Elliot, Steven Brem
Life of this patent
Estimating the term…
Abstract
Edema invariant tractography methods are provided. The methods include acquiring data using a multishell, high angular resolution diffusion imaging sequence; (ii) modeling the data using a multi-compartment model; and (iii) reconstructing the fibers through a probabilistic tracking algorithm using a fiber orientation distribution which has streamlines fitted to the output of said probabilistic tracking algorithm.
Claims34
An edema invariant tractography method comprising, with a processor:
- (i) acquiring data from a patient suspected of having edema using a 3-shell, high angular resolution diffusion imaging sequence, wherein:
- the first shell has a b-value of about 300 s/mm2 and a gradient direction of about 15, the second shell has a b-value of about 800 s/mm2 and a gradient direction of about 30, and the third shell has a b-value of about 2000 s/mm2 and a gradient direction of about 64;
- (ii) modeling said data using a multi-compartment model;
- and
- (iii) reconstructing fibers through a probabilistic tracking algorithm using a fiber orientation distribution which has streamlines fitted to the output of said probabilistic tracking algorithm,
- wherein the method is invariant to the presence of increased free water in the patient and provides fiber tracking in either the presence or absence of free water.
An edema invariant tractography method comprising, with a processor:
- (i) acquiring data from a patient suspected of having edema using a multishell, high angular resolution diffusion imaging sequence using b-values which are sensitive to free water diffusion and which provide enhanced angular resolution, wherein said multishell sequence comprises at least three shells, wherein a first shell has a b-value in the range of about 10 s/mm2 to about 400 s/mm2;
- (ii) modeling said data using a multi-compartment model comprising a first compartment which is a free water compartment which corrects for edema and a second compartment which models anisotropic diffusion within and around axons;
- and
- (iii) reconstructing fibers through a probabilistic tracking algorithm using a fiber orientation distribution which has streamlines fitted to the output of said probabilistic tracking algorithm
- wherein the method is invariant to the presence of increased free water in the patient and provides fiber tracking in either the presence or absence of free water.
A method of identifying tracts in an edematous brain of a patient, said method comprising, with a processor:
- (i) acquiring data from said patient using a 3-shell, high angular resolution diffusion imaging sequence, wherein:
- the first shell as a b-value of about 300 s/mm2 and a gradient direction of about 15, the second shell has a b-value of about 800 s/mm2 and a gradient direction of about 30, and the third shell has a b-value of about 2000 s/mm2 and a gradient direction of about 64;
- (ii) modeling said data using a multi-compartment model;
- and
- (iii) reconstructing fibers through a probabilistic tracking algorithm using a fiber orientation distribution which has streamlines fitted to the output of said probabilistic tracking algorithm, wherein the method is invariant to the presence of increased free water in the patient and provides fiber tracking in either the presence or absence of free water.
The method according to 24, wherein a tumor is present in said brain.
A method of identifying tracts in an edematous brain of a patient, said method comprising, with a processor:
- (i) acquiring data from the patient using a multishell, high angular resolution diffusion imaging sequence using b-values which are sensitive to free water diffusion and which provide enhanced angular resolution, wherein said multishell sequence comprises at least three shells, wherein a first shell has a b-value in the range of about 10 s/mm2 to about 400 s/mm2;
- (ii) modeling said data using a multi-compartment model comprising a first compartment which is a free water compartment which corrects for edema and a second compartment which models anisotropic diffusion within and around axons;
- (iii) reconstructing fibers through a probabilistic tracking algorithm using a fiber orientation distribution which has streamlines fitted to the output of said probabilistic tracking algorithm, wherein the method is invariant to the presence of increased free water in the patient and provides fiber tracking in either the presence or absence of free water.
A tractography electronic device, said device comprising at least one processing unit configured to:
- acquire data from a patient suspected of having edema using a multishell, high angular resolution diffusion imaging sequence using b-values which are sensitive to free water diffusion and which provide enhanced angular resolution, wherein said multishell sequence comprises at least three shells, wherein a first shell has a b-value in the range of about 10 s/mm2 to about 400 s/mm2;
- model said data using a multi-compartment model comprising a free water compartment;
- and
- reconstruct fibers through a probabilistic tracking algorithm using a fiber orientation distribution which has streamlines fitted to the output of said probabilistic tracking algorithm, wherein the product provides invariant fiber tracking in either the presence or absence of free water.
A computer program product, stored on a computer readable medium, said product comprising instructions for configuring a computer to:
- acquire data from a patient suspected of having edema using a 3-shell, high angular resolution diffusion imaging sequence, wherein:
- the first shell as a b-value of about 300 s/mm2 and a gradient direction of about 15, the second shell has a b-value of about 800 s/mm2 and a gradient direction of about 30, and the third shell has a b-value of about 2000 s/mm2 and a gradient direction of about 64;
- model said data using a multi-compartment model;
- and
- reconstruct fibers through a probabilistic tracking algorithm using a fiber orientation distribution which has streamlines fitted to the output of said probabilistic tracking algorithm, wherein the product provides invariant fiber tracking in either the presence or absence of free water.
Dependencies are inferred from the claim text.