Method and apparatus for detecting seizures

1. A system of apparatuses for monitoring a patient for seizure activity, the system comprising:

one or more electromyography electrodes configured to provide an electromyography signal;

a processor configured to receive the electromyography signal and process the electromyography signal to detect bursts characteristic of seizure activity based on whether regions of the electromyography signal meet criteria suitable to be qualified as bursts, including a comparison of criteria values to thresholds;

wherein said criteria values include a duration width and one or more of a signal-to-noise ratio and an amplitude;

wherein said thresholds include a minimum duration width, a maximum duration width, and one or more of a minimum signal-to-noise ratio, minimum amplitude, and maximum amplitude; and

wherein said processor is configured to determine a detected burst count and include said detected burst count in a determination of whether to send an alert indicating detection of said seizure activity to one or more remote devices.

2. The system of claim 1 wherein said processor is further configured to assign certainty weight values to bursts included among said detected burst count.

3. The system of claim 1 wherein said processor is further configured to determine a periodicity contribution based on a regularity of time periods between bursts and include said periodicity contribution in the determination whether to send said alert.

4. The system of claim 1 further comprising a base station in communication with said processor, wherein said one or more remote devices include said base station;

wherein said one or more electromyography electrodes and said processor are packaged as a single unit mountable to said patient and physically separate from said base station; and

wherein said base station is configured for receiving the alert and a transmitted portion of the electromyography signal, independently processing the transmitted portion of the electromyography signal to determine whether a seizure may be occurring, and generating an alarm if a seizure is detected.

5. The system of claim 4 wherein the base station is configured to select one or more of a plurality of pre-generated templates including base station threshold settings to determine whether a seizure is detected.

6. The system of claim 4 wherein the base station is configured to select base station threshold settings used to determine whether a seizure is detected that are based upon a location of the patient.

7. The system of claim 4 further comprising one or more environmental transceivers; and

wherein said one or more environmental transceivers are configured to transmit to said base station information about a location of the patient.

8. The system of claim 1 wherein said processor is further configured to wirelessly transmit a compressed form of burst data to a base station; and

wherein said base station is configured to use the compressed form of burst data to determine whether a seizure is detected using pre-defined base station threshold settings.

9. A method for monitoring a patient for seizure activity, the method comprising:

collecting an electromyography signal from the patient using one or more electromyography electrodes;

processing with one or more processors the electromyography signal to detect bursts characteristic of seizure activity based on whether regions of the electromyography signal meet criteria suitable to be qualified as said bursts, including a comparison of criteria values to thresholds;

wherein said criteria values include a duration width and one or more of a signal-to-noise ratio and an amplitude;

wherein said thresholds include a minimum duration width, a maximum duration width, and one or more of a minimum signal-to-noise ratio, minimum amplitude, and maximum amplitude;

determining a detected burst count; and

weighting said detected burst count in determining whether to send an alert indicating detection of said seizure activity to one or more remote devices.

10. The method of claim 9 further comprising assigning certainty weight values to bursts including among said detected burst count.

11. The method of claim 9 further comprising determining a periodicity contribution based on a regularity of time periods between bursts and weighting said periodicity contribution together with the weighting of said detected burst count in determining whether to send said alert.

12. The method of claim 11 wherein the periodicity contribution negatively weights the determination of whether to send said alert when time periods between bursts are too regular to be indicative of seizure activity.

13. A method of monitoring a patient for seizure activity comprising:

collecting an electromyography signal from the patient using one or more electromyography electrodes;

processing with a processor the electromyography signal to detect bursts, assign certainty values to individual bursts among said detected bursts, and determine a burst count contribution to seizure detection based on a number of said detected bursts weighted as a function of the certainty values assigned to said individual bursts;

wherein said certainty values are based on how well one or more characteristics of the individual bursts compare to one or more reference burst characteristics;

said one or more characteristics selected from a group of characteristics including burst signal-to-noise ratio, burst duration width, and burst amplitude;

wherein burst detection includes determining the presence of bursts based on whether regions of the electromyography signal meet criteria suitable to be qualified as bursts, including a comparison of criteria values to thresholds;

wherein said criteria values include a duration width and one or more of a signal-to-noise ratio and an amplitude;

wherein said thresholds include a minimum duration width, a maximum duration width, and one or more of a minimum signal-to-noise ratio, minimum amplitude, and maximum amplitude;

including said burst count contribution into an algorithm to determine whether a seizure is occurring; and

initiating an alert if a seizure is occurring.

14. The method of claim 13 further comprising identifying the presence of a plurality of bursts over a time period, determining a periodicity of bursts over said time period, determining a periodicity contribution to seizure detection, and including said periodicity contribution in determining whether a seizure is detected;

monitoring a location of the patient based on a detected location of the patient using one or more environmental transceivers; and

wherein said thresholds are included in a template file selected from among a plurality of pre-generated template files, the template file selected based on a patient location as determined by the monitoring of the patient location using said one or more environmental transceivers.

15. A method for reviewing patient electromyography data, the method comprising:

downloading from a computer memory an electromyography signal, the signal collected using one or more electromyography electrodes disposed on a patient while monitoring the patient during one or more monitoring periods;

processing with a processor the electromyography signal to detect bursts, assign certainty values to individual bursts among said detected bursts, and determine a burst count contribution to seizure detection based on a number of said detected bursts weighted as a function of the certainty values assigned to said individual bursts;

wherein said certainty values are based on how well one or more characteristics of the individual bursts compare to one or more reference burst characteristics;

said one or more characteristics selected from a group of characteristics including burst signal-to-noise ratio, burst duration width, and burst amplitude;

wherein burst detection includes determining the presence of bursts based on whether regions of the electromyography signal meet criteria suitable to be qualified as bursts, including a comparison of criteria values to thresholds;

wherein said criteria values include a duration width and one or more of a signal-to-noise ratio and an amplitude;

wherein said thresholds include a minimum duration width, a maximum duration width, and one or more of a minimum signal-to-noise ratio, minimum amplitude, and maximum amplitude;

including said burst count contribution into an algorithm to determine whether a seizure is occurring; and

initiating an alert if a seizure is occurring.