METHOD AND DEVICE FOR THE REAL-TIME MONITORING AND EVALUATION OF THE STATE OF A PATIENT WITH A NEUROLOGICAL CONDITION

Abstract

The invention relates to a method for the real-time monitoring and evaluation of the state of a patient with a neurological condition from the indicative parameters of the state of the patient obtained by means of an EEG, comprising a step of measuring the EEG by means of at least one sensor, a step of processing the measured values, a step of extracting at least one set of values of the indicative parameters of the state of the patient, each set of values being extracted from time segments of the EEG, a step of calculating for each time segment the risk level of the patient suffering a crisis due to the neurological condition, the risk level being calculated by applying at least one mathematical classification model to each corresponding set of values, and a step of classifying the state of the patient between at least an alert or preictal state and a non-alert or non-preictal state, depending on a threshold level.

Claims

1. A method for the real-time monitoring and evaluation of the state of a patient with a neurological condition from the indicative parameters (pi) of the state of the patient obtained by means of an EEG, characterised in that it comprises: a step of measuring the EEG (1) of the patient by means of at least one EEG sensor (S1, S2), a step of processing (2) the measured values of the EEG, a step of extracting (3) at least one set of values (x, y) of the indicative parameters (pi) of the state of the patient, each set of values (x, y) being extracted from different time segments (Tx, Ty) of the EEG, a step of calculating (4) for each time segment (Tx, Ty) the risk level (rp.sub.x, rp.sub.y) of the patient suffering a crisis due to the neurological condition, the risk level (rp.sub.x, rp.sub.y) being calculated by applying at least one mathematical classification model (m, n) to each corresponding set of values (x, y), and a step of classifying (5) the state of the patient between at least an alert or preictal state (A) and a non-alert or non-preictal state (B), depending on a predefined threshold level (u), and whereby when the risk level (rpx, rp.sub.y) in at least one time segment (Tx, Ty) is greater than the threshold level (u), the state of the patient is classified as preictal state (A), and when it is lower it is classified as non-preictal state (B).

2. The method according to claim 1, characterised in that more than one mathematical classification model (m, n) is applied to the sets of values (x, y) of each time segment (Tx, Ty) and the obtained risk level values of each mathematical classification model (m, n) are weighted with previously defined weights (pm, pn) so as to obtain a weighted risk level value (rpx, rp.sub.y) for each time segment (Tx, Ty).

3. The method according to claim 1, characterised in that at least one applied mathematical classification model (m, n) is among those known as SVM, LSBoost, Random Forest, KNN, Neural Networks, Naive Bayes, Gaussian process and ANN.

4. The method according to claim 1, characterised in that the sets of values (x, y) are extracted from time segments (Tx, Ty) overlapping one another in an overlap window between 20% and 60% of the time.

5. (canceled)

6. (canceled)

7. The method according to claim 1, characterised in that a warning signal is transmitted to the patient when 2 or more consecutive time segments (Tx) are classified as preictal (A).

8. The method according to claim 1, characterised in that a warning signal is transmitted to the patient when, in a group of between 3 to 30 consecutive segments, 3 or more time segments (Tx) are classified as preictal (A).

9. The method according to claim 1, characterised in that a warning signal is transmitted to the patient when a time segment (Tx) is classified as preictal (A).

10. The method according to claim 7, characterised in that between 10 to 30 time segments (Tx) after transmitting the warning signal to the patient, a query is transmitted whereby the patient must confirm if he or she has suffered a crisis.

11. (canceled)

12. (canceled)

13. The method according to claim 1, characterised in that more than one EEG sensor (S.sub.1, S.sub.2) measures the EEG of the patient, and wherein the steps of processing (2), extracting (3) and calculating (4) by each EEG sensor (S.sub.1, S.sub.2) are duplicated, the step of classifying (5) being performed for all the obtained sets of values (x, y).

14. (canceled)

15. (canceled)

16. (canceled)

17. The method according to claim 1, characterised in that it comprises a prior process of selecting (PS) the indicative parameters (pi), which comprises: a prior step of measuring (1′) the previous EEG of the patient, said EEG comprising sufficient previous time segments (Tx′, Ty′, Tz′) so that all the possible states of the condition of the patient are recorded, a prior step of processing (2′a) the measured values of the EEG, a prior step of calculating descriptive parameters (2′b) relative to the previous EEG, and a prior step of applying a method of selecting variables (2′c) to said descriptive parameters.

18. (canceled)

19. The method according to claim 17, characterised in that it comprises a prior process of selecting factors, said factors to be selected corresponding to: the mathematical classification models (m, n) to be applied in the step of calculating (4), the values of internal configuration parameters (pic) of said models (m, n) to be applied, the weights (p.sub.m, p.sub.n) in the event of applying more than one model (m,n), and the threshold level (u) to be applied in the step of classifying (5).

20. The method according to claim 19, characterised in that said process of selecting factors comprises the next steps: a prior step of extracting (3′) sets of previous values (x′, y′, z′) relative to the indicative parameters (pi) of each previous time segment (Tx′, Ty′, Tz′), establishing a first matrix (m1) with all the possible combinations for each: possible mathematical classification model (m′, n′) applicable to each set of previous values (x′, y′, z′), from a list of known models (m′, n′), possible combination of values of the internal configuration parameters (pic′) of each possible model (m′, n′), calculating a second matrix (m2) of possible risk levels (r′) for each set of previous values (x′, y′, z′) by means of each possible model (m′, n′) and for each possible combination of the first matrix (m1), establishing a third matrix (m3) with possible combinations of weights (p.sub.m′, p.sub.n′) of each possible model (m′, n′), calculating a fourth matrix (m4) of possible weighted risk levels (rp′) from: the possible risk levels (r′) of the second matrix (m2) the possible combinations of weights (p.sub.m′, p.sub.n′) of the third matrix (m3), establishing a fifth matrix (m5) of possible threshold risk levels (u′), obtaining a sixth matrix (m6) of classification of the hypothetical states of the patient, corresponding to the comparison of each possible weighted risk level (rp′) of the fourth matrix (m4) with each possible threshold risk level (u′) of the fifth matrix (m5), wherein said comparisons are designated as a hypothetical alert state (A′) when a possible weighted risk level (rp′) is equal to or exceeds the corresponding possible threshold risk level (u′) and as a hypothetical non-alert state (B′) when it does not exceed same, obtaining a seventh matrix (m7) of classification of each combination of factors, corresponding to the comparison of said hypothetical states (A′, B′) with real alert state (AR) and real non-alert state (BR) of the patient, the corresponding combination of factors being classified as a hit (OK) when the hypothetical and real states coincide, and as a miss (NOK) when they do not coincide, selecting the combination of factors obtaining a higher score in relation to a pre-established score depending on the hits and misses of the seventh matrix (m7).

21. (canceled)

22. A device for the real-time monitoring and evaluation of the state of a patient with a neurological condition, suitable for carrying out the described method according to claim 1, characterised in that it comprises: at least one sensor (S.sub.1, S.sub.2) located in an intra-auricular body (11) and configured for being in direct contact with the inside of an ear of the patient for measuring the EEG, and an electronics board in electrical contact with the sensor (S.sub.1, S.sub.2) and configured for carrying out the described method according to claim 1, including at least one processing unit and one wireless communication unit configured for communicating with a smartphone-type portable device.

23. (canceled)

24. The device according to claim 22, characterised in that it comprises a shell (12) housing the electronics board, and a flexible body (13) attaching the shell (12) to the intra-auricular body (11), said flexible body integrating therein the connection cables between the electronics board and the sensor (S.sub.1, S.sub.2).

25. The device according to claim 24, characterised in that the attachment between the intra-auricular body (11) and the flexible body (13) comprises a ball joint allowing rotation and spherical movement with respect to one another, while at the same time allowing electrical contact between the sensor (S.sub.1, S.sub.2) and the cables coming from the electronics board.

26. (canceled)

27. (canceled)

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0093] FIG. 1 illustrates an example of a measurement of an electroencephalogram signal with overlapping time segments, where the x-axis represents the time in seconds.

[0094] FIG. 2 illustrates the flow chart of a possible implementation of the present method for the real-time monitoring and evaluation of the state of a patient, showing on one side a possible method of selecting variables (PS) and on the other a possible implementation of the method for the real-time monitoring (PM). The discontinuous lines refer to the data transferred from one step to another.

[0095] FIG. 3 illustrates an embodiment of the monitoring device of the present invention.

BRIEF DESCRIPTION OF A PREFERRED EMBODIMENT

[0096] In view of the mentioned drawings and according to the numbering used, the figures show a preferred embodiment of the invention, which comprises the parts and elements indicated and described in detail below.

[0097] FIG. 1 shows an example of a double measurement of the electroencephalogram (EEG) signal of a patient, obtained with a reference electrode (E.sub.R) and two intra-auricular sensors (S.sub.1, S.sub.2) detecting brain activity from two different locations in the patient's ear. This figure shows the division of the values of the EEG into time segments (Tx′, Ty′, Tz′, Tx, Ty), on one side two segments (Tx, Ty) relative to the monitoring method (PM), and on the other side three previous time segments (Tx′, Ty′, Tz′) relative to the initial method of selecting the variables (PS), of 60 seconds each and exhibiting an overlap with the next one of 50%.

[0098] As is shown in FIG. 2, the preferred embodiment of the present method of monitoring the state of a patient with a neurological condition comprises an initial method of selecting the variables (PS), depending on the biology of each patient, and it serves to customise and optimise the subsequent monitoring method (PM).

[0099] This method of selecting variables (PS) comprises a first process of selecting indicative parameters (pi) including: [0100] a prior step of measuring (1′) the previous EEG of the patient, said EEG comprising sufficient previous time segments (TX′, TY′) so that all the possible states of the condition of the patient are recorded, [0101] a prior step of processing (2′a) the measured values of the EEG, [0102] a step of calculating descriptive parameters (2′b) relative to the previous EEG, and [0103] a step of applying a method of selecting variables (2′c) to said descriptive parameters for selecting the indicative parameters (pi).

[0104] In this preferred embodiment, the applied method of selecting variables is the Neighbourhood Component Analysis.

[0105] Moreover, in this embodiment carried out with two EEG sensors (S.sub.1, S.sub.2), all the previous steps are performed in parallel for both obtained measurements. The selected indicative parameters (pi) will be the resultant in any of the two measurements, although they are not repeated in both.

[0106] Next, the method of selecting variables (PS) comprises a process of selecting factors (m, n, pic, pm, pn, u) from said indicative parameters (pi), which correspond to: [0107] the mathematical classification models (m, n) to be applied in the step of calculating (two are indicated in this exemplary embodiment), [0108] the values of the internal configuration parameters (pic) of said models to be applied, [0109] the weights (pm, pn) in the event of applying more than one mathematical classification model (m, n), and [0110] the threshold level (u) to be applied in the step of classifying.

[0111] The process of selecting factors comprises a prior step of extracting (3′) several sets of values (x′, y′, z′) of the indicative parameters (pi) from each previous time segment (Tx′, Ty′, Tz′), and then comprises the following steps relative to the search for the multidimensional space with the best combinations of factors: [0112] establishing a first matrix (m1) with all the possible combinations of values of the internal configuration parameters (pic′) of each possible model (m′, n′) applicable to each set of previous values (x′, y′, z′), [0113] calculating a second matrix (m2) of possible risk levels (r′) for each set of previous values (x′, y′) by means of each possible model (m′, n′) and for each possible combination of the first matrix (m1), [0114] establishing a third matrix (m3) with possible combinations of weights (pm′, pn′) of each possible model (m′, n′), [0115] calculating a fourth matrix (m4) of possible weighted risk levels (rp′) from: [0116] the possible risk levels (r′) of the second matrix (m2) [0117] the possible combinations of weights (pm′, pn′) of the third matrix (m3), [0118] establishing a fifth matrix (m5) of possible threshold risk levels (u′), [0119] obtaining a sixth matrix (m6) of classification of the hypothetical states of the patient, corresponding to the comparison of each possible weighted risk level (rp′) of the fourth matrix (m4) with each possible threshold risk level (u′) of the fifth matrix (m5), wherein said comparisons are designated as a hypothetical alert state (A′) when a possible weighted risk level (rp′) is equal to or exceeds the corresponding possible threshold risk level (U′) and as a hypothetical non-alert state (B′) when it does not exceed same. [0120] obtaining a seventh matrix (m7) of classification of each combination of factors, corresponding to the comparison of said hypothetical states (A′, B′) with real alert state (AR) and real non-alert state (BR) of the patient, the corresponding combination of factors being classified as a hit when the hypothetical and real states coincide, and as a miss when they do not coincide,

[0121] Once this last matrix (m7) has been obtained, the combination of factors obtaining a higher score in relation to the hits and misses of the seventh matrix (m7) is selected from the following hierarchical order of values: [0122] the ratio of real preictal states that are detected with respect to those that are not detected, [0123] the ratio of real non-preictal states that are detected with respect to those that are not detected, [0124] the ratio of hypothetical preictal states that are hits with respect to those that are misses, [0125] the ratio of hypothetical non-preictal states that are hits with respect to those that are misses, [0126] the ratio of hits of preictal and non-preictal states balanced with respect to the sample size of each state, [0127] the ratio of preictal states that are hits and misses with respect to the total samples, [0128] the ratio of non-preictal states that are hits and misses with respect to the total samples, and [0129] the amount of previous consecutive segments (TX′, TY′) classified as non-preictal.

[0130] Once all the variables of the present monitoring method, i.e., both of the indicative parameters (pi) and of the factors (pic, pm, pn, u), have been selected, it is now possible to apply them to the monitoring method (PM).

[0131] As is shown in FIG. 2, the preferred embodiment of the monitoring method (PM) initially comprises a step of measuring the EEG (1) of the patient and a step of processing (2) the measured values. Next, for this preferred embodiment, a step of extracting (3) two sets of values (x, y) of indicative parameters (pi) is performed. Each set of values (x, y) corresponds to the values of a time segment (Tx, Ty) of the EEG contained in 60 seconds, the two time segments (Tx, Ty) overlapping by 50% in this preferred embodiment. Given that the device of this preferred embodiment has two EEG sensors (S.sub.1, S.sub.2), the preceding steps are performed in duplicate, i.e., by each sensor (S.sub.1, S.sub.2). In this case, the two sets of values (x, y) with which the method continues are the mean value of the values previously extracted in duplicate.

[0132] The next step corresponds to the step of calculating (4) the risk level of the patient suffering a crisis due to the neurological condition, for each time segment (Tx, Ty) and by means of each mathematical classification model (m, n). By way of example, this calculation can be performed by applying the mathematical classification models (m, n) called SVM and LSBoost to each set of values (x, y). The obtained risk level values for each mathematical classification model (m, n) are weighted with previously defined weights (pm, Pn,) so as to obtain a single weighted risk level value (rp.sub.x, rp.sub.y) for each time segment (Tx, Ty).

[0133] Next, there is a step of classifying (5) the state of the patient in each time segment (Tx, Ty), being classified in an alert or preictal state (A) when the weighted risk level (rp.sub.x, rp.sub.y) is greater than the previously defined threshold level (u), and a non-alert or non-preictal state (B) when it is lower.

[0134] In this preferred embodiment, the method comprises a step of sending (6) the classification of the state of the patient to a smartphone-type external device. In the event that a time segment (Tx, Ty) has been classified as preictal, the external device transmits a warning signal to the patient and to the person watching over them. In turn, after transmitting the warning signal, a query is transmitted whereby one of the two must confirm if the patient has suffered a crisis.

[0135] Once the state of the patient has been classified in a time segment (Tx, Ty), all the steps of the method are repeated again with the following time segments in order to re-classify the state of the patient.

[0136] FIG. 3 shows an embodiment of the device for the real-time monitoring of the state of a patient with a neurological condition, which is suitable for carrying out the method described above. This auricular device comprises a reference electrode (E.sub.R) and two sensors (S.sub.1, S.sub.2), configured for measuring the EEG, all of being of the silver/silver chloride type and conveniently located in an intra-auricular body (11) so that they can be in direct contact with the right ear of the patient. The figure shows one of the sensors (S.sub.1) and the reference electrode (E.sub.R) with dotted lines because they are located on the non-visible face of the intra-auricular body (11). In turn, the device comprises an electronics board (not seen in the figure), which includes a processing unit configured for processing the measurements obtained by the sensors (S.sub.1, S.sub.2) and carrying out the method described above, as well as a wireless communication unit configured for communicating with a device outside the auricular device.

[0137] In this preferred embodiment, the device also comprises a shell (12) which houses the electronics board, and a flexible body (13) which attaches the shell (12) to the intra-auricular body (11), said flexible body (13) integrating therein the connection cables between the electronics board and the sensors (S.sub.1, S.sub.2). In turn, this embodiment comprises an adapter for adapting the position (14) between the flexible body (13) and the intra-auricular body (11), provided with a cylindrical body (14a) through which there passes a rod (14b) attached to the flexible body (13), said rod being configured for moving and rotating with respect to the cylindrical body (14a), without being able to be completely removed. At the same time, the adapter (14) allows electrical contact between the sensors (S.sub.1, S.sub.2) and the cables coming from the electronics board. In this case, the attachment between the intra-auricular body (11) and the adapter for adapting the position (14) comprises a ball joint (not seen in the figure) allowing rotation and spherical movement with respect to one another, while at the same time allowing electrical contact between the sensors (S.sub.1, S.sub.2) and the cables coming from the electronics board.

[0138] The details, shapes, dimensions and other accessory elements, as well as the materials used in the manufacture of the present monitoring device may be suitably substituted with others that are technically equivalent and do not depart from the essential features of the invention or from the scope defined by the claims included below.