Method to identify optimum coronary artery disease treatment
A method of identifying an optimum treatment for a patient suffering from coronary artery disease, comprising: (i) providing patient information selected from: (a) status in the patient of one or more coronary disease associated biomarkers; (b) one or more items of medical history information selected from prior condition history, intervention history and medication history; (c) one or more items of diagnostic history, if the patient has a diagnostic history; and (d) one or more items of demographic data; (ii) aggregating the patient information in: (a) a Bayesian network; (b) a machine learning and neural network; (c) a rule-based system; and (d) a regression-based system; (iii) deriving a predicted probabilistic adverse event outcome for each intervention comprising percutaneous coronary intervention by placement of a bare metal stent, or a drug-coated stent; or by coronary artery bypass grafting; and (iv) determining the intervention having the lowest predicted probabilistic adverse outcome.
1. A method of identifying an optimum treatment for a patient suffering from coronary artery disease, the method comprising:
(i) providing one or more items of patient information selected from:
(A) status in the patient of one or more biomarkers associated with coronary heart disease;
(B) one or more items of medical history information of the patient selected from prior condition history, medical intervention history, and medication history;
(C) one or more items of diagnostic history of the patient; and
(D) one or more items of patient demographic data;
(ii) aggregating the patient information in a machine learning and neural network; and
(iii) determining a recommended intervention for the patient using a cascaded series of tests based on the aggregated patient information by:
deriving, using the machine learning and neural network based on a first set of patient information input to the machine learning and neural network, a predicted probability of target vessel revascularization (TVR) for a bare metal stent percutaneous coronary intervention;
in response to a determination that the predicted probability of TVR for the bare metal stent percutaneous coronary intervention is less than a first threshold, determining the bare metal stent percutaneous coronary intervention to be the recommended intervention for the patient;
in response to a determination that the predicted probability of TVR for the bare metal stent percutaneous coronary intervention is not less than the first threshold, deriving, using the machine learning and neural network based on a second set of patient information input to the machine learning and neural network, a predicted probability of a major adverse cardiac event (MACE) for a drug-eluting stent percutaneous coronary intervention, wherein the second set of patient information includes patient age, patient gender, diabetes condition of the patient, diagnostic history information of the patient, and CRP, CREA, DbTNT, parallel TNT, HsTNT and NTproBNP biomarkers;
in response to a determination that the predicted probability of a MACE for the drug-eluting stent percutaneous coronary intervention is not greater than a second threshold, determining the drug-eluting stent percutaneous coronary intervention to be the recommended intervention for the patient; and
in response to a determination that the predicted probability of a MACE for the drug-eluting stent percutaneous coronary intervention is greater than the second threshold, determining a coronary artery bypass grafting to be the recommended intervention for the patient,
wherein the machine learning and neural network models relationships among the first set of patient information and the predicted probability of TVR and among the second set of patient information and the predicted probability of a MACE without input from an expert regarding the relationships.
2. The method of claim 1 , wherein the first set of patient information includes a patient age, patient gender, a diabetes condition of the patient, diagnostic history information of the patient, and CREA, CK, CKMB, troponin, and CRP biomarkers.
3. A method of treating a patient suffering from coronary artery disease, the method comprising:
(i) providing one or more items of patient information selected from:
(A) status in the patient of one or more biomarkers associated with coronary heart disease;
(B) one or more items of medical history information of the patient selected from prior condition history, medical intervention history, and medication history;
(C) one or more items of diagnostic history of the patient; and
(D) one or more items of patient demographic data;
(ii) aggregating the patient information in a machine learning and neural network;
(iii) determining a recommended intervention for the patient using a cascaded series of tests based on the aggregated patient information by:
deriving, using the machine learning and neural network based on a first set of patient information input to the machine learning and neural network, a predicted probability of target vessel revascularization (TVR) for a bare metal stent percutaneous coronary intervention, wherein the first set of patient information includes patient age, patient gender, diabetes condition of the patient, diagnostic history information of the patient, and CREA, CK, CKMB, troponin and CRP biomarkers;
in response to a determination that the predicted probability of TVR for the bare metal stent percutaneous coronary intervention is less than a first threshold, determining the bare metal stent percutaneous coronary intervention to be the recommended intervention for the patient;
in response to a determination that the predicted probability of TVR for the bare metal stent percutaneous coronary intervention is not less than the first threshold, deriving, using the machine learning and neural network based on a second set of patient information input to the machine learning and neural network, a predicted probability of a major adverse cardiac event (MACE) for a drug-eluting stent percutaneous coronary intervention;
in response to a determination that the predicted probability of a MACE for the drug-eluting stent percutaneous coronary intervention is not greater than a second threshold, determining the drug-eluting stent percutaneous coronary intervention to be the recommended intervention for the patient; and
in response to a determination that the predicted probability of a MACE for the drug-eluting stent percutaneous coronary intervention is greater than the second threshold, determining a coronary artery bypass grafting to be the recommended intervention for the patient; and
(iv) initiating treatment of the patient with the recommended intervention determined for the patient,
wherein the machine learning and neural network models relationships among the first set of patient information and the predicted probability of TVR and among the second set of patient information and the predicted probability of a MACE without input from an expert regarding the relationships.
4. The method of claim 3 , wherein the biomarkers associated with coronary heart disease comprise one or more biomarkers selected from PIIINP, Tenascin-X, TIMP-1, MMP-3, MMP-9, fibrinogen, D-dimer, Activated protein C: C-inhibitor complex, tPA, IL-6, IL-1m IL-2, TNF, CRP, osteopontin, resistin, leptin, adiponectin, sCD28, sCD86, sCTLA-4, sVCAM-1, sICAM-1, endothelin-1, endothelin-2, HDL, LDL, lipoprotein-A, and apolipoprotein NB.
5. The method of claim 3 , wherein the first set of patient information input to the machine learning and neural network is selected to provide a maximum negative predictive value for the probability of TVR for the bare metal stent percutaneous coronary intervention in a set of training data from previous patients.
6. The method of claim 3 , wherein the second set of aggregated patient information input to the machine learning and neural network is optimized to provide a maximum positive predictive value for the probability of a MACE for the drug-eluting stent percutaneous coronary intervention in a set of training data from previous patients.
7. A method of identifying an optimum treatment for a patient suffering from coronary artery disease, the method comprising:
(i) providing one or more items of patient information selected from:
(A) status in the patient of one or more biomarkers associated with coronary heart disease;
(B) one or more items of medical history information of the patient selected from prior condition history, medical intervention history, and medication history;
(C) one or more items of diagnostic history of the patient; and
(D) one or more items of patient demographic data;
(ii) aggregating the patient information in a machine learning and neural network; and
(iii) determining a recommended intervention for the patient using a cascaded series of tests based on the aggregated patient information by:
deriving, using the machine learning and neural network based on a first set of patient information input to the machine learning and neural network, a predicted probability of target vessel revascularization (TVR) for a bare metal stent percutaneous coronary intervention, wherein the first set of patient information includes patient age, patient gender, diabetes condition of the patient, diagnostic history information of the patient, and CREA, CK, CKMB, troponin and CRP biomarkers;
in response to a determination that the predicted probability of TVR for the bare metal stent percutaneous coronary intervention is less than a first threshold, determining the bare metal stent percutaneous coronary intervention to be the recommended intervention for the patient;
in response to a determination that the predicted probability of TVR for the bare metal stent percutaneous coronary intervention is not less than the first threshold, deriving, using the machine learning and neural network based on a second set of patient information input to the machine learning and neural network, a predicted probability of a major adverse cardiac event (MACE) for a drug-eluting stent percutaneous coronary intervention;
in response to a determination that the predicted probability of a MACE for the drug-eluting stent percutaneous coronary intervention is not greater than a second threshold, determining the drug-eluting stent percutaneous coronary intervention to be the recommended intervention for the patient; and
in response to a determination that the predicted probability of a MACE for the drug-eluting stent percutaneous coronary intervention is greater than the second threshold, determining a coronary artery bypass grafting to be the recommended intervention for the patient,
wherein the machine learning and neural network models relationships among the first set of patient information and the predicted probability of TVR and among the second set of patient information and the predicted probability of a MACE without input from an expert regarding the relationships.
8. The method of claim 7 , wherein the biomarkers associated with coronary heart disease comprise one or more biomarkers selected from PIIINP, Tenascin-X, TIMP-1, MMP-3, MMP-9, fibrinogen, D-dimer, Activated protein C: C-inhibitor complex, tPA, IL-6, IL-1m IL-2, TNF, CRP, osteopontin, resistin, leptin, adiponectin, sCD28, sCD86, sCTLA-4, sVCAM-1, sICAM-1, endothelin-1, endothelin-2, HDL, LDL, lipoprotein-A, and apolipoprotein A/B.
9. The method of claim 7 , wherein the first set of patient information input to the machine learning and neural network is selected to provide a maximum negative predictive value for the predicted probability of TVR for the bare metal stent percutaneous coronary intervention in a set of training data from previous patients.
10. The method of claim 7 , wherein the second set of patient information input to the machine learning and neural network is optimized to provide a maximum positive predictive value for the predicted probability of a MACE for the drug-eluting stent percutaneous coronary intervention in a set of training data from previous patients.
11. The method of claim 7 , wherein the second set of patient information input to the machine learning and neural network includes a patient age, patient gender, a diabetes condition of the patient, diagnostic history information of the patient, and CRP, CREA, DbTNT, parallel TNT, HsTNT, and NTproBNP biomarkers.
12. The method of claim 7 , wherein a first set of biomarkers associated with coronary heart disease in the first set of patient information input to the machine learning and neural network is different from a second set of biomarkers associated with coronary heart disease in the second set of patient information input to the machine learning and neural network.
13. The method of claim 7 , further comprising:
deriving, using the machine learning and neural network, a predicted probability of a MACE for the bare metal stent percutaneous coronary intervention based on the aggregated patient information.
14. The method of claim 7 , further comprising:
deriving, using the machine learning and neural network, a predicted probability of TVR for the drug-eluting stent percutaneous coronary intervention based on the aggregated patient information.
15. The method of claim 7 , further comprising:
deriving, using the machine learning and neural network, a predicted probability of a MACE for the coronary artery bypass grafting based on the aggregated patient information.