Systems and methods for predicting location, onset, and/or change of coronary lesions
Systems and methods are disclosed for predicting the location, onset, or change of coronary lesions from factors like vessel geometry, physiology, and hemodynamics. One method includes: acquiring, for each of a plurality of individuals, a geometric model, blood flow characteristics, and plaque information for part of the individual's vascular system; training a machine learning algorithm based on the geometric models and blood flow characteristics for each of the plurality of individuals, and features predictive of the presence of plaque within the geometric models and blood flow characteristics of the plurality of individuals; acquiring, for a patient, a geometric model and blood flow characteristics for part of the patient's vascular system; and executing the machine learning algorithm on the patient's geometric model and blood flow characteristics to determine, based on the predictive features, plaque information of the patient for at least one point in the patient's geometric model.
1. A system for image processing, the system comprising:
a data storage device storing instructions for image processing; and
a processor configured to execute the instructions to perform a method comprising:
receiving time-varying images of a blood vessel with known plaque information at one or more points of the blood vessel;
receiving physiological and hemodynamic information at one or more points of the blood vessel for each of the received time-varying images with known plaque information;
determining changes in anatomical data over time at one or more points of the blood vessel for the received time-varying images;
creating feature vectors comprising: (1) features of the received physiological and hemodynamic information and (2) features of the determined changes in anatomical data, which are predictive of the plaque information of the blood vessel, at one or more points of the blood vessel for each of the received time-varying images with known plaque information;
associating the feature vectors with the plaque information at one or more points of the blood vessel for each of the received time-varying images;
training a machine learning algorithm for predicting plaque information from a patient at one or more points of a blood vessel, using the associated feature vectors;
receiving time-varying images of a blood vessel of a patient with unknown plaque information related to vascular hemodynamics of the blood vessel;
receiving physiological and hemodynamic information of the blood vessel of the patient for each of the received time-varying images with unknown plaque information;
determining changes in anatomical data over time for the received time-varying images of the blood vessel of the patient with unknown plaque information;
creating feature vectors comprising: (1) features of the received physiological and hemodynamic information and (2) features of the determined changes in anatomical data, which are predictive of the plaque information of the blood vessel, at one or more points of the blood vessel for each of the received time-varying images with unknown plaque information of the patient;
determining plaque information of the blood vessel of the patient, by inputting the created feature vectors for each of the received time-varying images with unknown plaque information of the patient into the trained machine learning algorithm; and
generating, using a computer processor, a patient-specific prediction or a patient-specific probability of a development of artery disease for the patient, using the determined plaque information.
2. The system according to claim 1 , wherein the processor is further configured for identifying a target region of the received time-varying images of the blood vessel of the patient to determine functional information.
3. The system according to claim 2 , wherein the processor is configured for: identifying the plaque information of the blood vessel of the patient by generating a geometric model; and
performing a blood flow simulation using computational flow dynamics.
4. The system according to claim 2 , wherein the processor is configured for:
determining changes in anatomical data over time with respect to the identified target region in the blood vessel region, on a basis of the received time-varying images;
generating a geometric model related to the identified target region, on the basis of the received time-varying images and the determined changes in anatomical data over time; and
analyzing the geometric model to determine (1) features of the received physiological and hemodynamic information and (2) features of the determined changes in anatomical data, which are predictive of the plaque information of the blood vessel, at one or more points of the blood vessel.
5. The system according to claim 1 , wherein the physiological and hemodynamic characteristics includes at least one of a cross-sectional area of the blood vessel.
6. The system according to claim 1 , wherein the processor is further configured for: mapping features of training data, which includes (1) the received physiological and hemodynamic information and (2) the determined changes in anatomical data that are predictive of the plaque information of the blood vessel, with the known plaque information at one or more points of the blood vessel.
7. The system according to claim 1 , wherein the processor is configured for: identifying plaque information of the blood vessel of the patient by further using a change in intensity value of the received time-varying images.
8. The system according to claim 1 , wherein the features of the received physiological and hemodynamic information and the determined changes in anatomical data that are predictive of the functional information includes features that are determined by a lumped parameter model.
9. The system according to claim 1 , wherein the change in anatomical data includes a change in blood vessel cross-sectional area and a change amount in the radius of the blood vessels.
10. The system according to claim 1 , wherein the processor is further configured for:
displaying information indicating the plaque information of the blood vessel of the patient;
calculating a change in cross-sectional area of the blood vessel of the patient, as the blood vessel morphology; and
displaying the change in cross-sectional area.
11. The system according to claim 10 , wherein the processor is configured for: displaying information indicating an electrocardiographic waveform of the patient during an image taking process and displaying the information indicating the change in anatomical data.
12. The system according to claim 11 , wherein the processor is configured for: displaying the information indicating the electrocardiographic waveform so as to be kept in correspondence with the information indicating the change in anatomical data.
13. The system according to claim 10 , wherein the processor is configured for: determining a local flow rate of blood flowing through the blood vessel of the patient.
14. The system according to claim 10 , wherein the processor is configured for: displaying a geometric model of the blood vessels of the patient and displaying the plaque information.
15. The method of claim 1 , wherein the plaque information comprises one or more of: a plaque location, onset, and/or change; an indication of a severity of a plaque; or an indication of an absence or presence of a plaque.
16. An image processing method comprising:
receiving time-varying images of a blood vessel with known plaque information at one or more points of the blood vessel related to vascular hemodynamics;
receiving physiological and hemodynamic information at one or more points of the blood vessel for each of the received time-varying images with known plaque information;
determining changes in anatomical data over time at one or more points of the blood vessel for the received time-varying images with known plaque information;
creating feature vectors comprising: (1) features of the received physiological and hemodynamic information and (2) features of the determined changes in anatomical data, which are predictive of the plaque information of the blood vessel, at one or more points of the blood vessel for each of the received time-varying images with known plaque information;
associating the feature vectors with the plaque information at one or more points of the blood vessel for each of the received time-varying images;
training a machine learning algorithm for predicting plaque information from a patient at one or more points of a blood vessel, using the associated feature vectors;
receiving time-varying images of a blood vessel of a patient with unknown plaque information related to vascular hemodynamics of the blood vessel;
receiving physiological and hemodynamic information of the blood vessel of the patient for each of the received time-varying images with unknown plaque information;
determining changes in anatomical data over time for the received time-varying images of the blood vessel of the patient with unknown plaque information;
creating feature vectors comprising: (1) features of the received physiological and hemodynamic information and (2) features of the determined changes in anatomical data, which are predictive of the plaque information of the blood vessel, at one or more points of the blood vessel for each of the received time-varying images with unknown plaque information of the patient;
determining plaque information of the blood vessel of the patient, by inputting the created feature vectors for each of the received time-varying images with unknown plaque information of the patient into the trained machine learning algorithm; and
generating, using a computer processor, a patient-specific prediction or a patient-specific probability of a development of artery disease for the patient, using the determined plaque information.
17. A non-transitory computer-readable storage medium having recorded thereon a plurality of computer-executable instructions that cause the computer to execute:
receiving time-varying images of a blood vessel with known plaque information at one or more points of the blood vessel related to vascular hemodynamics;
receiving physiological and hemodynamic information at one or more points of the blood vessel for each of the received time-varying images with known plaque information;
determining changes in anatomical data over time at one or more points of the blood vessel for the received time-varying images with known plaque information;
creating feature vectors comprising: (1) features of the received physiological and hemodynamic information and (2) features of the determined changes in anatomical data, which are predictive of the plaque information of the blood vessel, at one or more points of the blood vessel for each of the received time-varying images with known plaque information;
associating the feature vectors with the plaque information at one or more points of the blood vessel for each of the received time-varying images;
training a machine learning algorithm for predicting plaque information from a patient at one or more points of a blood vessel, using the associated feature vectors;
receiving time-varying images of a blood vessel of a patient with unknown plaque information related to vascular hemodynamics of the blood vessel;
receiving physiological and hemodynamic information of the blood vessel of the patient for each of the received time-varying images with unknown plaque information;
determining changes in anatomical data over time for the received time-varying images of the blood vessel of the patient with unknown plaque information;
creating feature vectors comprising: (1) features of the received physiological and hemodynamic information and (2) features of the determined changes in anatomical data, which are predictive of the plaque information of the blood vessel, at one or more points of the blood vessel for each of the received time-varying images with unknown plaque information of the patient;
determining plaque information of the blood vessel of the patient, by inputting the created feature vectors for each of the received time-varying images with unknown plaque information of the patient into the trained machine learning algorithm; and
generating, using a computer processor, a patient-specific prediction or a patient-specific probability of a development of artery disease for the patient, using the determined plaque information.