Method and Apparatus for Detecting Seizures Including Audio Characterization

1 . A method of detecting seizures with motor manifestations comprising:

receiving EMG data for a first period of time;

receiving audio data from said first period of time;

determining for said first period of time whether said EMG data meets a first EMG data threshold condition and/or if said audio data meets a first audio data threshold condition;

receiving EMG and audio data for a second period of time if either or both of said first EMG threshold condition and/or said first audio data threshold condition is met; and

determining for said second period of time whether either or both of said EMG data meets a second EMG data threshold condition and/or if said audio data meets a second audio data threshold condition;

initiating an alarm if, during said second time period, either or both of said second EMG threshold condition and/or said second audio data threshold condition is met.

2 . The method of claim 1 wherein meeting said first audio data threshold condition includes reaching a threshold level of audio signal amplitude followed by a sustained period of lower amplitude audio data.

3 . The method of claim 1 wherein meeting said first audio data threshold condition includes reaching an audio signal amplitude of at least about 50 decibels to about 75 decibels followed by a decreased audio signal, the decreased audio signal lasting for at least about 5 seconds.

4 . The method of claim 1 wherein meeting said first audio data threshold condition includes detection of one or more parts of audio data that repeat within a time period of about 0.2 to about 2 seconds.

5 . The method of claim 4 wherein said one or more parts of audio data that repeat are selected from a group of parts including a threshold amplitude of audio data, a threshold local maximum value in amplitude, a local maximum value in amplitude followed by a sustained decrease in amplitude of the audio data, and a data point in a pattern of audio data identified by pattern recognition.

6 . The method of claim 4 wherein said one or more parts of audio data that repeat include a portion of audio data qualified by regression analysis as being suitably similar to a model of portion of audio data.

7 . The method of claim 6 wherein said model portion of audio data is derived from recordings of patient's gasping for air during an inhalation part of a recorded seizure.

8 . The method of claim 4 wherein the one or more parts of audio data that repeat repeat at least about 4 to about 10 times to meet said first audio data threshold condition.

9 . The method of claim 1 wherein said second time period extends for a period of time of about 2 minutes from when said first threshold condition is met.

10 . (canceled)

11 . (canceled)

12 . (canceled)

13 . (canceled)

14 . (canceled)

15 . (canceled)

16 . (canceled)

17 . (canceled)

18 . (canceled)

19 . (canceled)

20 . (canceled)

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

receiving audio data and selecting from the received audio data a subset of audio data that may be indicative of a seizure;

transmitting the subset of audio data to a remote caregiver trained to interpret if the data is indicative of a seizure; and

triggering an alarm response if said audio data indicates that a seizure may be present.

22 . The method of claim 21 wherein said subset of audio data includes audio data identified by a pattern recognition program where an identified pattern repeats over a time period of about 0.2 to about 2 seconds, the identified pattern being present at least about 4 to about 10 times.

23 . The method of claim 21 further comprising detection of EMG signal data;

wherein said subset of audio data comprises data following detection of an increase in EMG signal amplitude.

24 . The method of claim 21 wherein the increase in EMG signal amplitude is an increase in EMG signal of about 2% to about 50% of a maximum voluntary contraction.

25 . A method of detecting seizures with motor manifestations comprising:

collecting audio data over a plurality of time periods using one or more acoustic sensors;

calculating one or more values of a characteristic of the collected acoustic data for each of a number of time periods among said plurality of time periods;

analyzing whether a value of the characteristic meets one or more criteria;

calculating one or more times between consecutive values that meet said one or more criteria;

determining whether said one or more times meet a periodicity condition for a patient experiencing a seizure; and

integrating the determination of periodicity in a decision about whether to initiate an alarm protocol.

26 . The method of claim 25 wherein said characteristic includes an acoustic amplitude; and

wherein said criteria includes whether said acoustic amplitude is a local maximum value that is greater than a threshold amplitude value.