Systems and methods for image processing to determine blood flow
Embodiments include systems and methods for determining cardiovascular information for a patient. A method includes receiving patient-specific data regarding a geometry of the patient's vasculature; creating an anatomic model representing at least a portion of the patient's vasculature based on the patient-specific data; and creating a computational model of a blood flow characteristic based on the anatomic model. The method also includes identifying one or more of an uncertain parameter, an uncertain clinical variable, and an uncertain geometry; modifying a probability model based on one or more of the identified uncertain parameter, uncertain clinical variable, or uncertain geometry; determining a blood flow characteristic within the patient's vasculature based on the anatomic model and the computational model of the blood flow characteristic of the patient's vasculature; and calculating, based on the probability model and the determined blood flow characteristic, a sensitivity of the determined fractional flow reserve to one or more of the identified uncertain parameter, uncertain clinical variable, or uncertain geometry.
1 . A system for processing angiographic image data to interactively segment blood vessels, wherein the system is configured to execute operations comprising:
at least one data storage device storing instructions for processing angiographic image data to interactively segment blood vessels; and
at least one processor configured to execute the instructions to perform operations comprising:
determining a three-dimensional geometric model based on image data of a patient;
determining one or more uncertain variables based at least in part on the three-dimensional geometric model and/or clinical data associated with the patient;
determining a range of at least one predicted blood flow characteristic value consistent with a predetermined confidence interval based on (i) simulating blood flow through the three-dimensional geometric model and the one or more uncertain variables, or based on (ii) applying a trained machine learning system to the three-dimensional geometric model and the one or more uncertain variables;
generating a visual representation of the three-dimensional geometric model that includes a visual indication of the determined range of the at least one predicted blood flow characteristic in relation to the predetermined confidence interval; and
outputting the visual representation.
2 . The system as defined in claim 1 , the operations further comprising:
determining a sensitivity of a location in the three-dimensional geometric model based on the one or more uncertain variables, wherein determining the sensitivity furthers comprises:
determining one or more locations of the three-dimensional geometric model as functionally significant locations based on a criterion that depends on the three-dimensional geometric model at the one or more locations, and
determining, given an uncertainty in a geometric characteristic of the three-dimensional geometric model, a value of the predicted blood flow characteristic at one or more of the functionally significant locations, wherein the determining further comprises determining the one of the functionally significant locations as a functionally significant location if the confidence interval value of the predicted blood flow characteristic, associated with the uncertainty of the three-dimensional geometric characteristic of the three-dimensional geometric model, is below a predetermined value at one or more of the functionally significant locations.
3 . The system as defined in claim 2 , wherein the criterion is based on determining a geometric characteristic of the three-dimensional geometric model at the one or more locations.
4 . The system as defined in claim 2 , wherein determining functionally significant locations further comprises determining one or more functionally significant locations where the value of the predicted blood flow characteristic at one or more of the functionally significant locations exceeds a predetermined threshold.
5 . The system as defined in claim 2 , wherein displaying the functionally significant locations comprises displaying the three-dimensional geometric model, wherein the functionally significant locations are indicated using visual indications.
6 . The system as defined in claim 2 , wherein the operations further comprise: allowing a user to modify the three-dimensional geometric model, at one or more of the functionally significant locations.
7 . The system as defined in claim 1 , wherein determining functionally significant locations further comprises: determining a value of the predicted blood flow characteristic at the functionally significant locations, and wherein displaying the functionally significant locations comprises displaying a value of an estimated fractional flow reserve (FFR) at the functionally significant locations.
8 . The system as defined in claim 7 , wherein determining the value of the estimated FFR comprises determining a confidence of the value of the estimated FFR at the functionally significant locations, and wherein displaying the functionally significant locations comprises displaying the confidence of the value of the estimated FFR at the functionally significant locations.
9 . The system as defined in claim 7 , wherein determining the value of the estimated FFR comprises determining a range of values of the estimated FFR at the functionally significant locations within a confidence.
10 . The system as defined in claim 1 , wherein the blood vessels include coronary arteries.
11 . A non-transitory computer readable medium for use on a computer system containing computer-executable programming instructions for processing angiographic image data to interactively segment blood vessels, the computer readable medium executing a method comprising:
determining a three-dimensional geometric model based on image data of a patient;
determining one or more uncertain variables based at least in part on the three-dimensional geometric model and/or clinical data associated with the patient;
determining a range of at least one predicted blood flow characteristic value consistent with a predetermined confidence interval based on (i) simulating blood flow through the three-dimensional geometric model and the one or more uncertain variables, or based on (ii) applying a trained machine learning system to the three-dimensional geometric model and the one or more uncertain variables;
generating a visual representation of the three-dimensional geometric model that includes a visual indication of the determined range of the at least one predicted blood flow characteristic in relation to the predetermined confidence interval; and
outputting the visual representation.
12 . A computer-implemented segmentation method for processing angiographic image data for interactively segmenting blood vessels, wherein the method comprises:
determining a three-dimensional geometric model based on image data of a patient;
determining one or more uncertain variables based at least in part on the three-dimensional geometric model and/or clinical data associated with the patient;
determining a range of at least one predicted blood flow characteristic value consistent with a predetermined confidence interval based on (i) simulating blood flow through the three-dimensional geometric model and the one or more uncertain variables, or based on (ii) applying a trained machine learning system to the three-dimensional geometric model and the one or more uncertain variables;
generating a visual representation of the three-dimensional geometric model that includes a visual indication of the determined range of the at least one predicted blood flow characteristic in relation to the predetermined confidence interval; and
outputting the visual representation.
13 . The method as defined in claim 12 , wherein the method further comprises:
determining a sensitivity of a location of the three-dimensional geometric model based on the one or more uncertain variables, wherein determining the sensitivity furthers comprises:
determining one or more locations of the three-dimensional geometric model as functionally significant locations based on a criterion that depends on the three-dimensional geometric model at the one or more locations; and
determining, given an uncertainty in a geometric characteristic of the three-dimensional geometric model, a value of the predicted blood flow characteristic at one or more of the functionally significant locations, wherein the determining further comprises determining the one of the functionally significant locations as a functionally significant location if the confidence interval value of the predicted blood flow characteristic, associated with the uncertainty of the three-dimensional geometric characteristic of the three-dimensional geometric model, is below a predetermined value at one or more of the functionally significant locations.
14 . The method as defined in claim 13 , wherein the criterion is based on determining a geometric characteristic of the three-dimensional geometric model at the one or more locations.
15 . The method as defined in claim 13 , wherein determining functionally significant locations further comprises determining one or more functionally significant locations where the value of the predicted blood flow characteristic at one or more of the functionally significant locations exceeds a predetermined threshold.
16 . The method as defined in claim 13 , wherein displaying the functionally significant locations comprises displaying the three-dimensional geometric model, wherein the functionally significant locations are indicated using visual indications.
17 . The method as defined in claim 12 , wherein determining functionally significant locations further comprises: determining a value of the predicted blood flow characteristic at the functionally significant locations, and wherein displaying the functionally significant locations comprises displaying a value of an estimated fractional flow reserve (FFR) at the functionally significant locations.
18 . The method as defined in claim 17 , wherein determining the value of the estimated FFR comprises determining a confidence of the value of the estimated FFR at the functionally significant locations, and wherein displaying the functionally significant locations comprises displaying the confidence of the value of the estimated FFR at the functionally significant locations.