Systems and methods for processing electronic images to predict 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 method for analyzing medical imaging data to assess a risk of coronary artery disease, comprising:
a) obtaining an image dataset comprising at least a portion of a heart of a patient;
b) segmenting at least a portion of a myocardium of the heart from the image dataset of a);
c) determining at least one geometrical feature of an inner vessel wall of at least one blood vessel surrounding the myocardium of the heart;
d) determining a value of at least one hemodynamic parameter based on the least one geometrical feature of c);
e) determining data segments characterizing the myocardium of the heart of b);
f) generating feature vectors related to the data segments of e), wherein the feature vectors include the at least one hemodynamic parameter of d);
g) classifying the data segments of e) by:
training a machine-learning classification model using training data regarding a presence or onset of plaque for various training data segments characterizing the myocardium of the heart of one or more individuals and (ii) associated training feature vectors generated based on the training data segments, such that the trained machine-learning model is trained to generate a respective patient-specific estimate of the presence or onset of plaque for each data segment in response to an input of a corresponding one of the feature vectors; and
for each of the data segments, inputting the corresponding one of the feature vectors of f) into the trained machine-learning classification model, such that the trained machine-learning model generates a respective patient-specific estimate of the presence or onset of plaque; and
h) providing, as an output, at least one of the respective patient-specific risk scores indicative estimate of the presence or onset of plaque for the patient based on the classifying of g).
2. The method of claim 1 , wherein:
the feature vectors of f) are generated based on patches of myocardial voxels and data that characterize the myocardium of the heart.
3. The method of claim 2 , wherein:
the data includes one or more of myocardium thickness, myocardium volume, ventricular volume and organ shape.
4. The method of claim 2 , wherein:
the myocardial voxels represent a coronary tree anatomy.
5. The method of claim 1 , wherein:
the training of the machine-learning classification model further causes the machine-learning classification model to learn a relationship between the training feature vectors and at least one reference parameter, wherein the at least one reference parameter is selected from the group consisting of:
i) an invasive fractional flow reserve measurement,
ii) a coronary flow reserve measurement,
iii) results of a cardiac stress test, and
iv) results of myocardial magnetic resonance imaging (MRI) perfusion, SPECT, PET, CT perfusion, or ultrasound.
6. The method of claim 1 , wherein:
the machine-learning classification model is based on a supervised machine learning algorithm.
7. The method of claim 6 , wherein:
the supervised machine learning algorithm is selected from a group consisting of a support vector machine, a neural network, a Bayesian classifier, and a Tree Ensemble.
8. The method of claim 1 , wherein:
the feature vectors of f) include data indicative of the myocardium of the heart.
9. The method of claim 1 , wherein:
the feature vectors of f) include additional information based on a patient ECG signal.
10. The method of claim 1 , wherein:
the at least one geometrical feature of c) represents a diameter or cross-section area of the inner vessel wall.
11. The method of claim 1 , wherein:
the at least one hemodynamic parameter of d) represents a pressure-drop or pressure gradient of a coronary vessel surrounding the myocardium.
12. The method of claim 1 , wherein:
the at least one hemodynamic parameter of d) represents at least one hemodynamic index of a coronary vessel surrounding the myocardium.
13. The method of claim 12 , wherein:
the at least one hemodynamic index relates to fractional flow reserve, coronary flow reserve, instantaneous wave-free ratio, hyperemic myocardium perfusion, index of microcirculatory resistance or other hemodynamic parameter along a coronary vessel surrounding the myocardium.
14. The method of claim 1 , wherein:
the feature vectors of f) represents variation in a characteristic of interest over myocardial voxels; and/or
the feature vectors of f) are derived from factors representing a characteristic of interest over multiple segments of the myocardium of the heart; and/or
the feature vectors of f) include additional information selected from the group consisting of: i) information that characterizes global features of the myocardium of the heart, ii) patient demographic information, iii) information that characterizes a presence or amount or type of coronary artery calcification or plaque, iv) information characterizing tissue composition or tissue type or contrast agent, v) information characterizing myocardium layering or myocardium layer, vi) information characterizing an ECG signal parameter, vii) information pertaining to a cardiac biomarker in blood such as cardiac troponin or creatine kinase, viii) information that characterizes fat surrounding the heart or inside the heart, ix) information that characterizes shape of the myocardium, x) information that characterizes myocardial strain, xi) information that characterizes blood volume or blood pressure, xii) information that characterizes cardiac output, and xiii) information that characterizes parts of the heart, ascending aortic or coronary tree.
15. The method of claim 1 , wherein:
operations a) to h) are performed by at least one processor.
16. The method of claim 1 , wherein the respective patient-specific estimate is location-specific.
17. A method for analyzing medical imaging data to assess a risk of coronary artery disease, comprising:
a) obtaining an image dataset comprising at least a portion of a heart of a patient;
b) segmenting at least a portion of a myocardium of the heart from image dataset of a);
c) determining at least one geometrical feature of an inner vessel wall of at least one blood vessel surrounding the myocardium of the heart;
d) determining a value of at least one hemodynamic parameter based on the least one geometrical feature of c);
e) determining data segments characterizing the myocardium of the heart of b);
f) generating feature vectors related to the data segments of e) based on myocardial voxels and data that characterize the myocardium of the heart;
g) classifying the data segments of e) by:
training a machine-learning classification model using training data regarding a presence or onset of plaque for various training data segments characterizing the myocardium of the heart of one or more individuals and (ii) associated training feature vectors generated based on the training data segments, such that the trained machine-learning model is trained to generate a respective patient-specific estimate of the presence or onset of plaque for each data segment in response to an input of a corresponding one of the feature vectors; and
for each of the data segments, inputting the corresponding one of the feature vectors of f) into the trained machine-learning classification model, such that the trained machine-learning model generates a respective patient-specific estimate of the presence or onset of plaque; and
h) providing, as an output, at least one of the respective patient-specific estimate of the presence or onset of plaque for the patient based on the classifying of g).
18. The method of claim 17 , wherein the respective patient-specific estimate is location-specific.
19. A non-transitory computer-readable medium storing instructions that, when executed by a processor, cause the processor to perform operations for analyzing medical imaging data to assess a risk of coronary artery disease, the operations comprising:
a) obtaining an image dataset comprising at least a portion of a heart of a patient;
b) segmenting at least a portion of a myocardium of the heart from the image dataset of a);
c) determining at least one geometrical feature of an inner vessel wall of at least one blood vessel surrounding the myocardium of the heart;
d) determining a value of at least one hemodynamic parameter based on the least one geometrical feature of c);
e) determining data segments characterizing the myocardium of the heart of b);
f) generating feature vectors related to the data segments of e), wherein the feature vectors include the at least one hemodynamic parameter of d);
g) classifying the data segments of e) by:
training a machine-learning classification model using training data regarding a presence or onset of plaque for various training data segments characterizing the myocardium of the heart of one or more individuals and (ii associated training feature vectors generated based on the training data segments, such that the trained machine-learning model is trained to generate a respective patient-specific estimate of the presence or onset of plaque for each data segment in response to an input of a corresponding one of the feature vectors; and
for each of the data segments, inputting the corresponding one of the feature vectors of f) into the trained machine-learning classification model, such that the trained machine-learning model generates a respective patient-specific estimate of the presence or onset of plaque; and
h) providing, as an output, at least one of the respective patient-specific estimate indicative of the presence or onset of plaque for the patient based on the classifying of g).