Certains contenus de cette application ne sont pas disponibles pour le moment.
Si cette situation persiste, veuillez nous contacter àObservations et contact
1. (WO2019002324) ESTIMATION DE VALEUR DE PARAMÈTRE DANS UN SYSTÈME DE COAGULATION
Note: Texte fondé sur des processus automatiques de reconnaissance optique de caractères. Seule la version PDF a une valeur juridique

CLAIMS:

1. A method for estimating a value of a parameter of a blood coagulation system in a subject that is periodically supplied with an anticoagulant, wherein the parameter of the coagulation system is a time taken for the coagulation system to reach a state of coagulation after the periodic supply of anticoagulant to the subject is discontinued, wherein the value of the parameter of the coagulation system is estimated based on concentration levels of a set of proteins in a sample derived from the coagulation system at a predetermined point in time relative to a point in time of supply of a last dose of anticoagulant to the coagulation system; wherein the method comprises the steps of:

receiving the concentration levels of the coagulation proteins derived from the coagulation system at the predetermined point in time relative to the point in time of supply of the last dose of anticoagulant to the coagulation system,

estimating the value of the parameter of the coagulation system based on the received concentration levels of the coagulation proteins using an estimating algorithm.

2. The method according to claim 1, wherein the set of proteins comprises 5 or 6 coagulation proteins.

3. The method according to any preceding claim, wherein the estimating algorithm is trained by a machine learning method.

4. The method according to any of claims 1 to 3, wherein the estimating algorithm is generated by an iterative method.

5. The method according to any of claims 1 to 4, wherein the estimating algorithm is generated by an estimating algorithm generation method comprising the steps:

initiating the estimating algorithm generation method with an estimated estimating algorithm and values of parameters of a test coagulation system,

calculating cross validation errors of the estimated estimating algorithm based on values of one or more of the parameters of the test coagulation system, and

improving the estimated estimating algorithm by selecting parameters to be used by the estimated estimating algorithm based on the calculated cross validation errors, wherein the steps of calculating cross validation errors and improving the estimated estimating algorithm are repeated until the occurrence of a predetermined event, and wherein the estimated estimating algorithm generated in the last iteration step of the estimating algorithm generation method is the estimating algorithm.

6. The method according to claim 5, wherein the step of initiating the estimating algorithm generation method with an estimated estimating algorithm and values of parameters of test coagulation systems comprises the steps:

providing an estimated estimating algorithm,

receiving target values from the test coagulation systems,

receiving input values of 5 or more parameters of each of the test coagulation systems, and

storing the received input values in a test parameter pool, wherein the input values are derived from the test coagulation systems and wherein the 5 or more parameters comprise the 5 or 6 coagulation proteins, and

wherein the step calculating cross validation errors of the estimated estimating algorithm based on values of one or more of the parameters of the test coagulation systems comprises the steps: moving one of the parameters of the test parameter pool to an estimation parameter pool of the estimated estimating algorithm,

improving the estimated estimating algorithm based on the parameters from the estimation parameter pool using logistic regression on the estimated estimating algorithm in a leave one out cross validation, wherein estimated values are estimated by the estimated estimating algorithm and

calculating a cross-validation error between the estimated values and the target values of the test coagulation systems, and

moving the one of the parameters back from the estimation parameter pool to the test parameter pool,

wherein the step of moving one of the parameters of the test parameter pool to the estimation parameter pool of the estimated estimating algorithm, improving the estimated estimating algorithm based on the parameters from the estimation parameter pool using logistic regression on the estimating algorithm in a leave one out cross validation, calculating a cross-validation error between the estimated values and the target values of the test

coagulation systems, and moving the one of the parameters back from the estimation parameter pool to the test parameter pool are repeated until all parameters of the test parameter pool have been moved once to the estimation parameter pool, and

wherein the step of improving the estimated estimating algorithm by selecting parameters to be used by the estimated estimating algorithm based on the calculated cross validation errors comprises the step:

moving the parameter with the lowest cross-validation error between the estimated values and the target values of the test coagulation systems permanently to the estimation parameter pool.

7. The method according to claim 2, wherein the 5 or 6 coagulation proteins are coagulation proteins that are affected by the anticoagulant, that affect the coagulation process of the coagulation system, or that are affected by the anticoagulant and affect the coagulation process of the coagulation system.

8. The method according to claim 2, wherein 5 of the 5 or 6 coagulation proteins are coagulation factor VII (F7), coagulation factor IX (F9), Anti-thrombin (AT), coagulation factor X (F10), and coagulation factor VIII (F8).

9. The method according to claim 7 or 8, wherein one of the 5 or 6 coagulation proteins is coagulation factor XI (Fl 1).

10. The method according to claim 3, wherein the estimating algorithm is generated by a supervised learning method based on (i) values from a plurality of subjects of a time taken for the coagulation system to reach a state of coagulation after the periodic supply of anticoagulant to each subject is discontinued, and (ii) concentration levels from the plurality of subjects of the coagulation proteins derived from the coagulation system at the predetermined point in time relative to the point in time of supply of the last dose of anticoagulant to the coagulation system of each subject.

1 1. The method according to any preceding claim, further comprising determining a point in time for discontinuing periodic supply of anticoagulant, to determine a point in time for a surgery based on the point in time for discontinuing periodic supply of

anticoagulant, or to provide the coagulation system with another drug.

12. The method according to any preceding claim, wherein the state of coagulation is indicated by an international normalized ratio (INR) value of below 1.5.

13. A decision support device (10) comprising:

a receiving unit (12) for receiving concentration levels of a set of proteins in a sample derived from a blood coagulation system (20) in a subject at a predetermined point in time relative to a point in time of supply of a last dose of anticoagulant to the coagulation system, and

a system unit (14) for estimating a value of a parameter of the coagulation system (20) using an estimating algorithm, wherein the parameter of the coagulation system is a time taken for the coagulation system to reach a state of coagulation after periodic supply of anticoagulant to the subject is discontinued,

wherein the receiving unit (12) is configured to supply received concentration levels of the proteins derived from the coagulation system (20) to the system unit (14), and wherein the system unit (14) is configured to estimate the value of the parameter of the coagulation system (20) based on concentration levels of the set of proteins using the estimating algorithm.

14. The device (10) according to claim 13 comprising a user interface (16), wherein the user interface (16) is configured to supply received concentration levels of the coagulation proteins derived from the coagulation system (20) to the system unit (14) and to receive the estimated value of the parameter of the coagulation system (20) from the system unit (14), and wherein the user interface (16) is configured to receive by a user (22) using the decision support device (10) as input to the decision support device (10) the concentration levels of the set of coagulation proteins of the coagulation system (20) and to provide to the user (22) using the decision support device (10) as output of the decision support device (10) the value of the parameter estimated by the system unit (14).

15. The method according to claim 13 or 14, wherein the set of proteins comprises

5 or 6 coagulation proteins.

16. A computer program for estimating a value of a parameter of a coagulation system (20), wherein the computer program comprises program code for causing a processor

to carry out the method as defined in any of claims 1 to 12, when the computer program is run on the processor.

17. Computer readable medium (24) comprising a computer program according to claim 16.