Plaque vulnerability assessment in medical imaging
Rather than rely on variation from physician to physician and limited imaging information for assessing plaque vulnerability of a patient, medical imaging and other information are used by a machine-implemented classifier to predict plaque rupture. Anatomical, morphological, hemodynamic, and biochemical features are used in combination to classify plaque.
1. A method for assessing plaque vulnerability of a patient in medical imaging, the method comprising:
extracting an anatomical feature of a vessel or plaque and a morphological feature of the plaque from a scan of the patient by a medical imaging scanner;
obtaining a hemodynamic feature of the patient from a hemodynamic sensor or personalized modeling of the vessel based on the scan;
receiving at an interface a biochemical feature from a blood test of the patient, the biochemical feature including a level of high sensitive troponin;
calculating by a machine-implemented classifier a risk score for plaque rupture for the patient and a risk stratification, both the risk score and risk stratification calculated by the machine-implemented classifier from the anatomical, morphological, hemodynamic, and biochemical features; and
transmitting the risk score for the patient to a display.
2. The method of claim 1 wherein extracting the morphological feature comprises extracting a type of the plaque from an angiography as the scan.
3. The method of claim 1 wherein extracting the anatomical feature comprises extracting a shape, geometry, inlet angle, outlet angle, location of the plaque relative to the vessel, stenosis of the vessel, vessel size, vessel shape, or combinations thereof.
4. The method of claim 1 wherein extracting comprises extracting automatically with a processor or semi-automatically with the processor and a user input.
5. The method of claim 1 wherein obtaining comprises obtaining a pressure, flow, or pressure and flow and calculating stress from the pressure, flow, or pressure and flow.
6. The method of claim 1 wherein the level of high sensitive troponin is a level of high sensitive troponin T.
7. The method of claim 1 wherein calculating comprises calculating by the machine-implemented classifier as a weighted sum of ranks mapped from each of the anatomical, morphological, hemodynamic, and biochemical features, the weights based on clinical literature.
8. The method of claim 1 wherein calculating comprises calculating by the machine-implemented classifier comprises a mechanistic model.
9. The method of claim 1 wherein calculating comprises calculating by the machine-implemented classifier comprising a machine-learnt classifier.
10. The method of claim 1 wherein calculating comprises calculating the risk score as part of the risk stratification and a plaque vulnerability assessment for the patient in a coronary artery disease analysis.
11. The method of claim 1 wherein calculating comprises calculating the risk score for the patient in an acute coronary syndrome analysis.
12. The method of claim 1 further comprising repeating extracting, obtaining, and calculating for another plaque.
13. The method of claim 1 further comprising performing the scan with the medical imaging scanner comprising a non-invasive modality, wherein the anatomical, morphological, and hemodynamic features are all obtained without invasion of the patient.
14. A method for assessing plaque vulnerability of a patient in medical imaging, the method comprising:
loading a morphological characteristic of plaque of a patient, a hemodynamic characteristic, an anatomy characteristic of a vessel with the plaque, and a biochemical characteristic of the plaque as an input feature vector, where a biochemical characteristic is a level of high sensitive troponin;
classifying a vulnerability of the plaque of the patient by a machine-trained classifier in response to the input feature vector, the classifying being part of a risk stratification based on the hemodynamic, biochemical, and anatomy characteristics; and
signaling a result of the classifying to a user.
15. A method for assessing plaque vulnerability of a patient in medical imaging, the method comprising:
extracting an anatomical feature of a vessel or plaque and a morphological feature of the plaque from a scan of the patient by a medical imaging scanner;
obtaining a hemodynamic feature of the patient from a hemodynamic sensor or personalized modeling of the vessel based on the scan;
receiving at an interface a biochemical feature from a blood test of the patient, the receiving comprising receiving a level of high sensitive troponin;
calculating by a machine-implemented classifier a risk score for plaque rupture for the patient and a risk stratification, both the risk score and risk stratification calculated by the machine-implemented classifier from the anatomical, morphological, hemodynamic, and biochemical features; and
transmitting the risk score for the patient to a display.