Algorithm for detecting a seizure from cardiac data

Patent metadata

Field	Value
Publication number	US-9241647-B2
Application number	US-201314059334-A
Country	US
Kind code	B2
Filing date	Oct 21, 2013
Priority date	Apr 29, 2010
Publication date	Jan 26, 2016
Grant date	Jan 26, 2016

Abstract

Methods, systems, and apparatus for detecting the seizure in a patient using a medical device. The determination is performed by collecting cardiac data determining valid heart beats suitable for seizure detection from the cardiac data; calculating heart rate data of interest from the valid heart beats; and identifying a seizure event from the heart rate data. The medical device may then take a responsive action, such as warning, logging the time of the seizure, computing and storing one or more seizure severity indices, and/or treating the seizure.

First claim

What is claimed:

1. A method for detecting epileptic seizures based upon a time of beat sequence of the patient's heart comprising:
collecting cardiac data from the patient's heart, said cardiac data comprising a time series of fiducial time markers for candidate heartbeats;
identifying valid beats from said candidate heartbeats by subjecting a plurality of candidate beats to at least one beat validity test, said at least one beat validity test comprising at least one beat interval test applied to a candidate beat interval derived from a candidate heartbeat and at least one preceding heartbeat;
accepting as valid beats the candidate beats that pass said at least one beat validity test;
forming a first window comprising a first valid beat and at least one prior beat, wherein said first window is selected from:
a time window of from 1 to 10 seconds, bounded on the most recent end by said first valid beat; and
a number of beats window comprising said first valid beat and a number of immediately preceding beats ranging from 2-15;

testing said first window with at least one window test, wherein said at least one window test comprises a dispersion test;
accepting the beats in said window as suitable for seizure detection if said window has a dispersion less than a threshold level of dispersion;
in response to accepting the beats in said window as suitable for seizure detection, determining a foreground heart rate parameter comprising a statistical measure of central tendency of heart rate in said first window;
forming a second window larger than said first window;
determining a background heart rate parameter comprising a statistical measure of central tendency of heart rate in said second window;
determining a relative heart rate comprising at least one of a first ratio of said foreground to said background heart rate parameters and a second ratio of said background to said foreground heart rate parameters;
comparing said relative heart rate to at least one of a first seizure threshold value associated with said first ratio and a second seizure threshold value associated with said second ratio;
detecting an epileptic seizure if at least one of said first ratio exceeds said first seizure threshold value, or said second ratio is below said second seizure threshold value; and
taking a responsive action based upon the detecting, wherein the responsive action is selected from providing a warning to the patient or a caregiver, notifying the patient or the caregiver, logging the time of occurrence of said epileptic seizure, storing one or more seizure severity indices relating to said epileptic seizure, or treating said epileptic seizure.

2. The method of claim 1 , wherein identifying valid beats comprises determining if each of said plurality of candidate beats falls within a plausibly physiological interval.

3. The method of claim 2 , wherein said beat validity test comprises comparing said candidate beat interval to at least one of an upper beat interval threshold and a lower beat interval threshold.

4. The method of claim 3 , wherein said upper and lower beat interval thresholds are derived from at least one of the patient's own heartbeat data, and heartbeat data from a sample patient population based upon one or more of brain state, sex, age, weight, level of activity, time of day, type of epilepsy, use of drugs or substances (such as food) that affect cardiac function, ambient temperature, body temperature, respiration, and blood pressure.

5. The method of claim 1 , wherein said at least one beat interval test comprises:
a) determining that said candidate beat interval corresponds to a heart rate within a range bound by a minimum heart rate and a maximum heart rate;
b) determining that the candidate beat interval is within an acceptable percentage of at least one of the immediately preceding valid beat interval or a recent baseline heart rate in a predetermined time window; and
c) determining that the absolute slope derived from the current candidate beat interval and the most recent beat interval does not correspond to a rate of change of heart rate that is physiologically improbable.

6. The method of claim 5 , wherein:
a) said minimum heart rate is about 35 beats per minute and said maximum heart rate is about 180 beats per minute;
b) said candidate beat interval is not more than about 115 percent of the greater of the immediately preceding valid beat interval or a recent baseline heart rate in a 30 second time window, and is at least about 65 percent of the immediately preceding valid beat interval; and
c) said absolute slope of the current candidate beat interval and the most recent beat interval is ≦0.3.

7. The method of claim 1 , wherein said at least one window test comprises:
a) determining whether the mean square error of a least squares linear fit of the beats in said first window is less than or equal to a predetermined heart rate variability threshold.

8. The method of claim 1 , wherein said first window comprises a time window and said at least one window test comprises determining that the number of valid beats in the window exceeds a lower number of beats threshold.

9. The method of claim 1 , wherein determining a foreground heart rate parameter comprises determining a target percentile value in a uniform distribution Percentile Tracking Filter in said first window.

10. The method of claim 9 , wherein upper and lower bounds for said uniform distribution are adaptively determined for said first window based upon the maximum and minimum beat intervals in said first window.

11. The method of claim 10 , wherein said target percent value of said Percentile Tracking Filter comprises a value in the range of 20 percent to 80 percent.

12. The method of claim 1 , wherein determining a background heart rate parameter comprises determining a target percentile value in a uniform distribution Percentile Tracking Filter in said second window.

13. The method of claim 12 , further comprising an exponential forgetting factor applied to the interbeat intervals of the heartbeats in said Percentile Tracking Filter.

14. The method of claim 1 , further comprising the step of determining a duration of time that
said first ratio exceeds said first seizure threshold value, or
said second ratio is below said second seizure threshold value, and wherein detecting an epileptic seizure occurs only if said duration exceeds a seizure duration threshold.

15. The method of claim 1 , wherein said dispersion test comprises determining the mean squared error of a least-squares linear fit of the heartbeats in the first window, and wherein said heartbeats in said window are accepted as suitable for seizure detection if said mean squared error is less than a threshold.

16. The method of claim 1 , wherein said second window is selected from:
a time window of from 10 seconds to 86,400 seconds; and
a number of beats window comprising from about 30 beats to about 175,000 beats.

17. A method for detecting epileptic seizures based upon a time of beat sequence of the patient's heart comprising:
collecting cardiac data from the patient's heart, said cardiac data comprising a time series of fiducial time markers for candidate heartbeats;
identifying valid beats from said candidate heartbeats by subjecting a plurality of candidate beats to at least one beat validity test, said at least one beat validity test comprising at least one beat interval test applied to a candidate beat interval derived from a candidate heartbeat and at least one preceding heartbeat;
accepting as valid beats

Assignees

• Flint Hills Scient Llc
• Cyberonics Inc

Inventors

• Osorio Ivan
• Frei Mark G

Classifications

• A61B5/0245

  by using sensing means generating electric signals, {i.e. ECG signals} · CPC title

• A61N1/36064

  Epilepsy · CPC title

• A61B5/04012 Primary

  Human Necessities · mapped topic

• A61B5/02405

  Determining heart rate variability · CPC title

• A61B5/4094 Primary

  Diagnosing or monitoring seizure diseases, e.g. epilepsy · CPC title