Systems and methods for predicting tissue viability deficits from physiological, anatomical, and patient characteristics

1. A computer-implemented method for estimating tissue viability of a patient's tissue, the method comprising:

receiving tissue viability data for a plurality of individuals;

receiving data of one or more physiological or anatomical parameters, for each of the plurality of individuals;

training, by using a training set comprising the tissue viability data and the data of one or more physiological or anatomical parameters, a machine learning algorithm, wherein the machine learning algorithm maps the one or more physiological or anatomical parameters to tissue viability;

receiving image data derived from one or more images of a patient's anatomy;

generating one or more patient-specific models, each being a model of a vessel or tissue of the patient, using the image data;

calculating patient-specific values of the one or more physiological or anatomical parameters based on the one or more patient-specific models;

computing a tissue viability value by inputting the patient-specific values of the one or more physiological or anatomical parameters into the trained machine learning algorithm, wherein the trained machine learning algorithm includes feature weights defining an extent to which the one or more physiological or anatomical parameters are predictive of tissue viability; and

outputting, to an electronic storage medium or a display, the computed tissue viability value or a treatment plan generated based on the computed tissue viability value.

2. The computer implemented method of claim 1 , wherein

the one or more physiological or anatomical parameters include perfusion territories,

the patient-specific values include a patient tissue perfusion territory estimation,

the method comprises:

estimating a blood supply to the one or more vessel or tissue areas, using a blood flow simulation in at least one of the one or more patient-specific models;

determining the patient tissue perfusion territory estimation based on the estimated blood supply;

modifying the patient-specific model or the patient-specific values of the one or more physiological or anatomical parameters;

determining an effect of the modifying the patient-specific model or at least one of the patient-specific values of the one or more physiological or anatomical parameters on the computed tissue viability value; and

generating a treatment plan based on the determined effect, and

outputting the treatment plan to the electronic storage medium or the display.

3. The computer implemented method of claim 2 , wherein the estimated blood supply is for a physiological state that includes a resting patient state, a hyperemic state, an exercise state, a postprandial state, a gravitational state, an emotional state, a state of hypertension, a medicated state, or a combination thereof.

4. The computer implemented method of claim 2 , wherein the estimated blood supply includes fractional flow reserve, flow magnitude, flow direction, or a combination thereof.

5. The computer implemented method of claim 2 , wherein the treatment plan is generated based further on the modified patient-specific model or the patient-specific values of the one or more physiological or anatomical parameters.

6. The computer implemented method of claim 1 , wherein the one or more physiological or anatomical parameters include one or more anatomical characteristics including vessel size, vessel shape, vessel tortuosity, vessel length, vessel thickness, or a combination thereof.

7. The computer implemented method of claim 1 , wherein the tissue viability value includes a measure of an extent to which a vessel, tissue, or organ is functional.

8. The computer implemented method of claim 1 , wherein:

the computing the tissue viability value further includes inputting, into the machine learning algorithm, one or more image characteristics of the tissue or vessel in one or more physiological states, wherein the one or more image characteristics are derived from the image data and includes, one or more of:

local average intensities,

texture characteristics, and

standard image features.

9. The computer implemented method of claim 1 , wherein the computing the tissue viability value further includes inputting, into the machine learning algorithm, one or more secondary characteristics including patient characteristics, target tissue disease characteristics, electromechanical measurements, or a combination thereof.

10. The computer implemented method of claim 1 , wherein the computing the tissue viability value further includes comparing blood flow characteristics in the tissue or a vessel at different physiological states.

11. The computer implemented method of claim 1 , wherein the one or more patient-specific models include:

a coronary vascular model and a model of the myocardium;

a cerebral vascular model and a model of the brain;

a peripheral vascular model and a model of muscle;

a hepatic vascular model and a model of a liver;

a renal vascular model and a model of a kidney;

a visceral vascular model and a model of a bowel; or

a vascular model representing a vessel and a target organ to which blood is supplied by the vessel.

12. The computer implemented method of claim 1 , further comprising:

adjusting the patient-specific values of the one or more physiological or anatomical parameters based on the computed tissue viability value; and

simulating a blood flow characteristic using the computed tissue viability value and the adjusted patient-specific values.

13. A system for estimating patient-specific tissue viability, the system comprising:

a data storage device storing instructions for determining characteristics of tissue viability; and

a processor configured to execute the instructions to perform a method including the steps of:

receiving tissue viability data for a plurality of individuals;

receiving data of one or more physiological or anatomical parameters, for each of the plurality of individuals;

training, by using a training set comprising the tissue viability data and the data of the one or more physiological or anatomical parameters, a machine learning algorithm, wherein the machine learning algorithm maps the one or more physiological or anatomical parameters to tissue viability;

receiving image data derived from one or more images of a patient's anatomy;

generating one or more patient-specific models, each being a model of a vessel or tissue of the patient, using the image data;

calculating patient-specific values of the one or more physiological or anatomical parameters based on the one or more patient-specific models;

computing a tissue viability value by inputting the patient-specific values of the one or more physiological or anatomical parameters into the trained machine learning algorithm, wherein the trained machine learning algorithm includes feature weights defining an extent to which the one or more physiological or anatomical parameters are predictive of tissue viability; and

outputting, to an electronic storage medium or a display, the computed tissue viability value or a treatment plan generated based on the computed tissue viability value.

14. The system of claim 13 , wherein

the one or more physiological or anatomical parameters include perfusion territories,

the patient-specific values include a patient tissue perfusion territory estimation,

the method further comprises:

estimating a blood supply to the one or more vessel or tissue areas, using a blood flow simulation in at least one of the one or more patient-specific models;

determining the patient tissue perfusion territory estimation based on the estimated blood supply;

modifying the patient-specific model or the patient-specific values of the one or more physiological or anatomical parameters;

determining an effect of the modifying the patient-specific model or at least one of the patient-specific values of the one or more physiological or anatomical parameters on the computed tissue viability value; and

generating a treatment plan based on the determined effect, and

outputting the treatment plan to the electronic storage medium or the display.

15. The system of claim 14 , wherein the estimated blood supply is for a physiological state that includes a resting patient state, a hyperemic state, an exercise state, a postprandial state, a gravitational state, an emotional state, a state of hypertension, a medicated state, or a combination thereof.

16. The system of claim 14 , wherein the estimated blood supply includes fractional flow reserve, flow magnitude, flow direction, or a combination thereof.

17. The system of claim 13 , wherein the tissue viability value includes a measure of the extent to which a vessel, tissue, or organ is functional.

18. The system of claim 13 , wherein the estimating the tissue viability value includes comparing blood flow characteristics in a tissue or a vessel at different physiological states.

19. The system of claim 13 , wherein the one or more patient-specific models include:

a coronary vascular model and the myocardium;

a cerebral vascular model and the brain;

a peripheral vascular model and muscle;

a hepatic vascular model and a liver;

a renal vascular model and a kidney;

a visceral vascular model and a bowel; or

a vascular model representing a vessel and a target organ to which blood is supplied by the vessel.

20. A non-transitory computer readable medium for performing a method on a computer system containing computer-executable programming instructions for estimating a characteristic of tissue viability, the method comprising:

receiving tissue viability data for a plurality of individuals;

receiving data of one or more physiological or anatomical parameters, for each of the plurality of individuals;

training, by using a training set comprising the tissue viability data and the data of the one or more physiological or anatomical parameters, a machine learning algorithm, wherein the machine learning algorithm maps the one or more physiological or anatomical parameters to tissue viability;

receiving image data derived from one or more images of a patient's anatomy;

generating one or more patient-specific models, each being a model of a vessel or tissue of the patient, using the image data;

calculating patient-specific values of the one or more physiological or anatomical parameters based on the one or more patient-specific models;

computing a tissue viability value by inputting the patient-specific values of the one or more physiological or anatomical parameters into the trained machine learning algorithm, wherein the trained machine learning algorithm includes feature weights defining an extent to which the one or more physiological or anatomical parameters are predictive of tissue viability; and

outputting, to an electronic storage medium or a display, the computed tissue viability value or a treatment plan generated based on the computed tissue viability value.