Systems and methods for seizure detection with a statistical analysis and an artificial intelligence analysis

1. A seizure detection system comprising one or more circuits,

wherein the one or more circuits are configured to:

receive an electroencephalogram (EEG) signal generated based on electrical brain activity of a patient;

identify candidate seizures with a statistical analysis that identifies the candidate seizures based on changes in non-linear features of the EEG signal;

cause a display device to display the candidate seizures identified by the statistical analysis;

train an artificial intelligence model to classify the candidate seizures based on training data comprising the EEG signal and the candidate seizures identified by the statistical analysis;

determine to switch from displaying the candidate seizures identified with the statistical analysis to displaying the candidate seizures identified with the artificial intelligence model responsive to a determination that a performance of the artificial intelligence model to classify the candidate seizures satisfies a threshold; and

switch from displaying the candidate seizures identified with the statistical analysis to displaying the candidate seizures identified based on the artificial intelligence model with the EEG signal.

2. The seizure detection system of claim 1 , wherein the one or more circuits are configured to:

determine a first performance level of the statistical analysis; and

determine a second performance level of the artificial intelligence model;

wherein the one or more circuits are configured to determine to switch from displaying the candidate seizures identified with the statistical analysis to displaying the candidate seizures identified with the artificial intelligence model in response to determining that the second performance level is greater than the first performance level.

3. The seizure detection system of claim 1 , wherein the one or more circuits are configured to determine to switch from displaying the candidate seizures identified with the statistical analysis to displaying the candidate seizures identified the artificial intelligence model by determining that at least a predefined amount of detections of the candidate seizures by the statistical analysis are also detected by the artificial intelligence model.

4. The seizure detection system of claim 1 , wherein the one or more circuits are configured to identify the candidate seizures with the statistical analysis by:

determining a plurality of metrics based on the EEG signal, the plurality of metrics indicating the non-linear features of the EEG signal; and

determining that the EEG signal indicates the candidate seizures by determining, based at least in part on the plurality of metrics, changes in the non-linear features of the EEG signal over time, wherein the changes in the non-linear features indicates physiological forces that give rise to the candidate seizures.

5. The seizure detection system of claim 4 , wherein determining, based at least in part on the plurality of metrics, the changes in the non-linear features of the EEG signal comprises determining an increase in the non-linear features over time.

6. The seizure detection system of claim 4 , wherein the plurality of metrics comprise at least one of dimensionality, synchrony, Lyapunov exponents, one or more forms of entropy, one or more forms of eigenvalues, global non-linearity, distance differences between recurrence trajectories, higher order spectra, loss of complexity, a surrogate test, or self-similarity.

7. The seizure detection system of claim 4 , wherein the one or more circuits are configured to determine that the EEG signal indicates the candidate seizures by:

determining, based at least in part on the plurality of metrics, the changes in the non-linear features of the EEG signal over time by:

performing a preliminary analysis with one metric of the plurality of metrics, wherein the preliminary analysis indicates that the EEG signal indicates a candidate seizure or that the EEG signal includes noise; and

performing a secondary analysis with one or more metrics of the plurality of metrics other than the one metric to determine whether the EEG signal indicates the candidate seizure or that the EEG signal includes the noise.

8. The seizure detection system of claim 1 , wherein the one or more circuits are configured to:

operate in a first operating phase where the one or more circuits identify the candidate seizures with the statistical analysis; and

operate in a second operating phase where the one or more circuits identify the candidate seizures with the artificial intelligence model.

9. The seizure detection system of claim 8 , wherein the one or more circuits are configured to operate in the first operating phase by:

generating seizure alerts indicating that the EEG signal indicates the candidate seizures identified by the statistical analysis; and

causing a user interface to display the seizure alerts.

10. The seizure detection system of claim 8 , wherein the one or more circuits are configured to operate in the first operating phase by:

training the artificial intelligence model based on training data while the candidate seizures are identified by the statistical analysis.

11. The seizure detection system of claim 10 , wherein the one or more circuits are configured to operate in the first operating phase by training the artificial intelligence model based on the EEG signal and the candidate seizures identified by the statistical analysis.

12. The seizure detection system of claim 10 , wherein the one or more circuits are configured to operate in the first operating phase by:

generating seizure alerts indicating that the EEG signal indicates the candidate seizures identified by the statistical analysis;

causing a user interface to display the seizure alerts;

receiving labels of the candidate seizures or the EEG signal from a user via the user interface; and

training the artificial intelligence model based on the labels and the EEG signal.

13. A seizure detection method comprising:

receiving, by a processing circuit, an electroencephalogram (EEG) signal generated based on electrical brain activity of a patient;

identifying, by the processing circuit, candidate seizures with a statistical analysis that identifies the candidate seizures based on changes in non-linear features of the EEG signal;

causing, by the processing circuit, a display device to display the candidate seizures identified by the statistical analysis;

training, by the processing circuit, an artificial intelligence model to classify the candidate seizures based on training data comprising the EEG signal and the candidate seizures identified by the statistical analysis;

determining, by the processing circuit, to switch from displaying the candidate seizures identified with the statistical analysis to displaying the candidate seizures identified with the artificial intelligence model responsive to a determination that a performance of the artificial intelligence model to classify the candidate seizures satisfies a threshold; and

switching, by the processing circuit, from displaying the candidate seizures identified with the statistical analysis to displaying the candidate seizures identified based on the artificial intelligence model with the EEG signal.

14. The method of claim 13 , further comprising:

determining, by the processing circuit, a first performance level of the statistical analysis; and

determining, by the processing circuit, a second performance level of the artificial intelligence model;

wherein determining, by the processing circuit, to switch from displaying the candidate seizures identified with the statistical analysis to displaying the candidate seizures identified with the artificial intelligence model in response to determining that the second performance level is greater than the first performance level.

15. The method of claim 13 , wherein determining, by the processing circuit, to switch from identifying the candidate seizures with the statistical analysis to the artificial intelligence model by determining that at least a predefined amount of detections of the candidate seizures by the statistical analysis are also detected by the artificial intelligence model.

16. The method of claim 13 , wherein identifying, by the processing circuit, the candidate seizures with the statistical analysis comprises:

determining a plurality of metrics based on the EEG signal, the plurality of metrics indicating the non-linear features of the EEG signal; and

determining that the EEG signal indicates the candidate seizures by determining, based at least in part on the plurality of metrics, changes in the non-linear features of the EEG signal over time, wherein the changes in the non-linear features indicates physiological forces that give rise to the candidate seizures.

17. The method of claim 13 , further comprising:

operating, by the processing circuit, in a first operating phase by identifying the candidate seizures with the statistical analysis; and

operating, by the processing circuit, in a second operating phase by identifying the candidate seizures with the artificial intelligence model.

18. The method of claim 17 , wherein operating, by the processing circuit, in the first operating phase comprises:

generating seizure alerts indicating that the EEG signal indicates the candidate seizures identified by the statistical analysis; and

causing a user interface to display the seizure alerts.

19. The method of claim 17 , wherein operating, by the processing circuit, in the first operating phase comprises:

training the artificial intelligence model based on training data while the candidate seizures are identified by the statistical analysis.

20. The method of claim 19 , wherein operating, by the processing circuit, in the first operating phase comprises training the artificial intelligence model based on the EEG signal and the candidate seizures identified by the statistical analysis.

21. The method of claim 19 , wherein operating, by the processing circuit, in the first operating phase comprises:

generating seizure alerts indicating that the EEG signal indicates the candidate seizures identified by the statistical analysis;

causing a user interface to display the seizure alerts;

receiving labels of the candidate seizures or the EEG signal from a user via the user interface; and

training the artificial intelligence model based on the labels and the EEG signal.

22. A seizure detection system comprising:

one or more electrodes connected to a patient, the one or more electrodes configured to generate an electroencephalogram (EEG) signal based on electrical brain activity of the patient; and

a processing circuit configured to:

receive the EEG signal from the one or more electrodes;

identify candidate seizures with a statistical analysis that identifies the candidate seizures based on changes in non-linear features of the EEG signal;

cause a display device to display the candidate seizures identified by the statistical analysis;

train an artificial intelligence model to classify the candidate seizures based on training data comprising the EEG signal and the candidate seizures identified by the statistical analysis;

determine to switch from displaying the candidate seizures identified with the statistical analysis to displaying the candidate seizures identified with the artificial intelligence model responsive to a determination that a performance of the artificial intelligence model to classify the candidate seizures satisfies a threshold; and

switch from displaying the candidate seizures identified with the statistical analysis to displaying the candidate seizures identified based on the artificial intelligence model with the EEG signal.

23. The seizure detection system of claim 22 , wherein the artificial intelligence model receives output from the statistical analysis and identifies the candidate seizures based at least in part on the output received from the statistical analysis.

24. The seizure detection system of claim 22 , wherein the artificial intelligence model includes an input and a weight applied to the input;

wherein the processing circuit is configured to:

provide a candidate seizure determination made by the statistical analysis to the input of the artificial intelligence model; and

provide a confidence level of the candidate seizure determination made by the statistical analysis as the weight of the artificial intelligence model.

25. The seizure detection system of claim 22 , wherein the processing circuit is configured to:

receive one or more indications of false positives indicating candidate seizure identifications incorrectly made by at least one of the statistical analysis or the artificial intelligence model; and

train the artificial intelligence model based on the one or more indications of the false positives.