Methods and systems for predicting sensitivity of blood flow calculations to changes in anatomical geometry

Abstract

Embodiments include methods and systems and for determining a sensitivity of a patient's blood flow characteristic to anatomical or geometrical uncertainty. For each of one or more of individuals, a sensitivity of a blood flow characteristic may be obtained for one or more uncertain parameters. An algorithm may be trained based on the sensitivities of the blood flow characteristic and one or more of the uncertain parameters for each of the plurality of individuals. A geometric model, a blood flow characteristic, and one or more of the uncertain parameters of at least part of the patient's vascular system may be obtained for a patient. The sensitivity of the patient's blood flow characteristic to one or more of the uncertain parameters may be calculated by executing the algorithm on the blood flow characteristic of at least part of the patient's vascular system, and one or more of the uncertain parameters.

1 - 20 . (canceled) 
     
     
         21 . A computer-implemented method of determining a sensitivity of a patient's blood flow characteristic to uncertainty in a geometric model of a patient's vascular system, the method comprising:
 obtaining, for a patient, a geometric model of at least a portion of a vascular system and determining at least one sensitivity of a blood flow characteristic to at least one uncertainty in geometry in the geometric model;   mapping, in a machine learning database, a plurality of features of the geometric model to the obtained sensitivities;   determining, for the patient, at least one value of uncertainty in geometry in the geometric model of a vessel region the patient's vascular system; and   determining a sensitivity of a blood flow characteristic of the patient to at least one value of uncertainty in geometry in the geometric model of at least part of the patient's vascular system, using the machine learning database and the plurality of features.   
     
     
         22 . The method of  claim 21 , wherein obtaining the at least one sensitivity of a blood flow characteristic to the at least one uncertainty in geometry in the geometric model comprises calculating the at least one sensitivity by assigning a probabilistic distribution function to the at least one uncertainty. 
     
     
         23 . The method of  claim 21 , wherein obtaining the at least one sensitivity of the blood flow characteristic to the at least one uncertainty in geometry in the geometric model comprises calculating the at least one sensitivity by solving a stochastic algorithm for a vessel region in the geometric model. 
     
     
         24 . The method of  claim 23 , wherein solving the stochastic algorithm comprises utilizing quadrature points identified by a Smolyak sparse grid algorithm. 
     
     
         25 . The method of  claim 23 , further comprising creating a decision tree using the mapped plurality of features and the obtained sensitivities in the machine learning database, and wherein determining the sensitivity of the blood flow characteristic of the patient to the at least one value of uncertainty further comprises using the decision tree. 
     
     
         26 . The method of  claim 21 , wherein determining the sensitivity of the blood flow characteristic of the patient comprises identifying a mapped feature among the plurality of features in the machine learning database. 
     
     
         27 . The method of  claim 21 , further comprising splitting the mapped plurality of features and the obtained sensitivities in the machine learning database into a training set and a test set. 
     
     
         28 . A system for determining a sensitivity of a patient's blood flow characteristic to uncertainty in a geometric model of a patient's vascular system, the system comprising:
 a data storage device storing instructions for determining sensitivity; and   a processor configured to execute the instructions to perform a method including the steps of:
 obtaining, for a patient, a geometric model of at least a portion of a vascular system and determining at least one sensitivity of a blood flow characteristic to at least one uncertainty in geometry in the geometric model; 
 mapping, in a machine learning database, a plurality of features of the geometric model to the obtained sensitivities; 
 obtaining, for the patient, at least one value of uncertainty in geometry in the geometric model of a vessel region the patient's vascular system; and 
 determining a sensitivity of a blood flow characteristic of the patient to at least one value of uncertainty in geometry in the geometric model of at least part of the patient's vascular system, using the machine learning database and the plurality of features. 
   
     
     
         29 . The system of  claim 28 , wherein obtaining the at least one sensitivity of a blood flow characteristic to the at least one uncertainty in geometry in the geometric model comprises calculating the at least one sensitivity by assigning a probabilistic distribution function to the at least one uncertainty. 
     
     
         30 . The system of  claim 28 , wherein obtaining the at least one sensitivity of the blood flow characteristic to the at least one uncertainty in geometry in the geometric model comprises calculating the at least one sensitivity by solving a stochastic algorithm for a vessel region in the geometric model. 
     
     
         31 . The system of  claim 30 , wherein solving the stochastic algorithm comprises utilizing quadrature points identified by a Smolyak sparse grid algorithm. 
     
     
         32 . The system of  claim 30 , wherein, for each geometric model, a resolution of the sensitivity may be increased by solving a stochastic algorithm for an increased number of vessel regions of the geometric model. 
     
     
         33 . The system of  claim 28 , wherein the processor is further configured to determine the sensitivity of the blood flow characteristic of the patient by identifying a mapped feature among the plurality of features in the machine learning database. 
     
     
         34 . The system of  claim 28 , wherein the processor is further configured to split the mapped plurality of features and the obtained sensitivities in the machine learning database into a training set and a test set. 
     
     
         35 . A non-transitory computer readable medium for use on at least one computer system containing computer-executable programming instructions for determining a sensitivity of a patient's blood flow characteristic to uncertainty in a geometric model of a patient's vascular system, the instructions comprising steps for:
 obtaining, for a patient, a geometric model of at least a portion of a vascular system and at least one sensitivity of a blood flow characteristic to at least one uncertainty in geometry in the geometric model;   mapping, in a machine learning database, a plurality of features of the geometric model to the obtained sensitivities;   determining, for the patient, at least one value of uncertainty in geometry in the geometric model of a vessel region of the patient's vascular system; and   determining a sensitivity of a blood flow characteristic of the patient to at least one value of uncertainty in geometry in the geometric model of at least part of the patient's vascular system, using the machine learning database and the plurality of features.   
     
     
         36 . The computer readable medium of  claim 35 , wherein obtaining the at least one sensitivity of a blood flow characteristic to the at least one uncertainty in geometry in the geometric model comprises calculating the at least one sensitivity by assigning a probabilistic distribution function to the at least one uncertainty. 
     
     
         37 . The computer readable medium of  claim 35 , wherein obtaining the at least one sensitivity of the blood flow characteristic to the at least one uncertainty in geometry in the geometric model comprises calculating the at least one sensitivity by solving a stochastic algorithm for a vessel region in the geometric model. 
     
     
         38 . The computer readable medium of  claim 37 , wherein solving the stochastic algorithm comprises utilizing quadrature points identified by a Smolyak sparse grid algorithm. 
     
     
         39 . The computer readable medium of  claim 37 , wherein, for each geometric model, a resolution of the at least one sensitivity may be increased by solving a stochastic algorithm for an increased number of vessel regions of the geometric model. 
     
     
         40 . The computer readable medium of  claim 35 , wherein determining the sensitivity of the blood flow characteristic of the patient comprises identifying a mapped feature among the identified features in the machine learning database.