Detecting, assessing and managing epilepsy using a multi-variate, metric-based classification analysis

• US 8,684,921 B2
• Filed: 05/15/2012
• Issued: 04/01/2014
• Est. Priority Date: 10/01/2010
• Status: Active Grant

• Current Assignee: Flint Hills Scientific, LLC
• Original Assignee: Flint Hills Scientific, LLC
• Inventors: Osorio, Ivan
• Primary Examiner(s): Gedeon, Brian T
• Application Number:
  US13/472,365
• Publication Number:
  US 20120310050A1
• Time in Patent Office: 686 Days
• Field of Search: 600/16, 600/17, 600/300, 600/483, 600/508, 600/509, 600515-519
• US Class Current: 600/300
• CPC Class Codes: A61B 2560/0242 adapted to measure environmental factors, e.g. temperature, pollution
• CPC Class Codes: A61B 2560/0475 Special features of memory means, e.g. removable memory cards
• CPC Class Codes: A61B 5/0205 Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• CPC Class Codes: A61B 5/02055 Simultaneously evaluating both cardiovascular condition and temperature
• CPC Class Codes: A61B 5/021 Measuring pressure in heart or blood vessels A61B5/0205 takes precedence
• CPC Class Codes: A61B 5/024 Detecting, measuring or recording pulse rate or heart rate A61B5/0205, A61B5/021 take precedence
• CPC Class Codes: A61B 5/02405 Determining heart rate variability
• CPC Class Codes: A61B 5/026 Measuring blood flow A61B3/1233, A61B3/1241 take precedence
• CPC Class Codes: A61B 5/08 Detecting, measuring or recording devices for evaluating the respiratory organs A61B5/0205 takes precedence
• CPC Class Codes: A61B 5/11 Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb for measuring pulse A61B5/02 ;A61B5/1038 takes precedence; motion detection to correct for motion artifacts in physiological signals A61B5/721
• CPC Class Codes: A61B 5/1117 Fall detection
• CPC Class Codes: A61B 5/1118 Determining activity level
• CPC Class Codes: A61B 5/14503 invasive, e.g. introduced into the body by a catheter or needle or using implanted sensors A61B5/1459, A61B5/1464, A61B5/1473, A61B5/1482, A61B5/14865 take precedence
• CPC Class Codes: A61B 5/14539 for measuring pH
• CPC Class Codes: A61B 5/14542 for measuring blood gases A61B5/14551 takes precedence
• CPC Class Codes: A61B 5/283 Invasive
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle
• CPC Class Codes: A61B 5/369 Electroencephalography [EEG] devices for psychotechnics A61B5/16
• CPC Class Codes: A61B 5/389 Electromyography [EMG]
• CPC Class Codes: A61B 5/4094 : Diagnosing or monitoring seizure diseases, e.g. epilepsy
• CPC Class Codes: A61B 5/4836 : Diagnosis combined with treatment in closed-loop systems or methods A61B5/0036 takes precedence
• CPC Class Codes: A61B 5/4848 : Monitoring or testing the effects of treatment, e.g. of medication
• CPC Class Codes: A61B 5/486 : Bio-feedback A61B5/375 takes precedence
• CPC Class Codes: A61B 5/686 : Permanently implanted devices, e.g. pacemakers, other stimulators, biochips A61B5/6861 takes precedence
• CPC Class Codes: A61B 5/7246 : using correlation, e.g. template matching or determination of similarity
• CPC Class Codes: A61B 5/7264 : Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/7267 : involving training the classification device
• CPC Class Codes: A61B 5/7275 : Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: A61B 5/7282 : Event detection, e.g. detecting unique waveforms indicative of a medical condition A61B5/4094, A61B5/4818, A61B5/349, A61B5/0823 take precedence
• CPC Class Codes: A61B 5/742 : using visual displays A61B5/339 takes precedence
• CPC Class Codes: A61B 5/743 : Displaying an image simultaneously with additional graphical information, e.g. symbols, charts, function plots
• CPC Class Codes: A61B 5/746 : Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms alarms per se G08B
• CPC Class Codes: A61M 16/1005 : with O2 features or with parameter measurement
• CPC Class Codes: A61M 2202/0208 : Oxygen
• CPC Class Codes: A61M 2205/3303 : Using a biosensor
• CPC Class Codes: A61N 1/36064 : Epilepsy
• CPC Class Codes: A61N 1/36135 : using physiological parameters
• CPC Class Codes: G16H 20/10 : relating to drugs or medications, e.g. for ensuring correct administration to patients
• CPC Class Codes: G16H 20/40 : relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture
• CPC Class Codes: G16H 50/20 : for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/30 : for calculating health indices; for individual health risk assessment

Abstract

A method for identifying changes in an epilepsy patient'"'"'s disease state, comprising: receiving at least one body data stream; determining at least one body index from the at least one body data stream; detecting a plurality of seizure events from the at least one body index; determining at least one seizure metric value for each seizure event; performing a first classification analysis of the plurality of seizure events from the at least one seizure metric value; detecting at least one additional seizure event from the at least one determined index; determining at least one seizure metric value for each additional seizure event, performing a second classification analysis of the plurality of seizure events and the at least one additional seizure event based upon the at least one seizure metric value; comparing the results of the first classification analysis and the second classification analysis; and performing a further action.

Claims

1. A method for identifying changes in an epilepsy patient'"'"'s disease state, comprising:

receiving at least one body data stream;

determining at least one of an autonomic index, a neurologic index, a metabolic index, an endocrine index, a tissue index, or a tissue stress index, a physical fitness or body integrity index based upon the at least one body data stream;

detecting a plurality of seizure events based upon the at least one determined index;

determining at least one seizure metric value for each seizure event in the plurality of seizure events;

performing a first classification analysis of the plurality of seizure events based on the at least one seizure metric value for each seizure event;

detecting at least one additional seizure event based upon the at least one determined index;

determining at least one seizure metric value for each of the at least one additional seizure events,

performing a second classification analysis of the plurality of seizure events and the at least one additional seizure event based upon the at least one seizure metric value;

comparing the results of the first classification analysis and the second classification analysis; and

performing a further action selected from;

a. reporting a change from the first classification to the second classification;

b. reporting the absence of a change from the first classification to the second classification;

c. displaying a result of at least one of the first classification analysis, the second classification analysis, and the comparing;

d. identifying the emergence of a new class based on the comparing;

e. identifying the disappearance of a prior class based on the comparing;

f. identifying one or more outlier seizure events not part of any class;

g. identifying an effect of a therapy;

h. providing a therapy to the patient in response to the comparing;

i. identifying a proposed change in therapy;

j. identifying a proposed additional therapy;

k. identifying an extreme seizure event;

l. issuing a warning if a new seizure class appears or an extreme event occurs;

m. logging to memory the time, date and type of change in the patient'"'"'s seizures.

2. The method of claim 1,

wherein performing a first classification analysis comprises identifying one or more seizure classes by determining one or more relationships among at least a portion of the plurality of seizure events, wherein the one or more relationships are based on the at least one seizure metric value for each seizure event;

wherein performing a second classification analysis comprises identifying one or more seizure classes by determining one or more relationships among at least a portion of the plurality of seizure events and the at least one additional seizure event, wherein the one or more relationships are based on the at least one seizure metric value for each seizure event; and

wherein comparing the results of the first classification analysis and the second classification analysis comprises identifying a change in at least one class of said one or more seizure classes from said first classification analysis to said second classification analysis.

3. The method of claim 2, wherein each of the one or more seizure classes identified in the first classification analysis and the second classification analysis comprises a seizure cluster based upon the at least one seizure metric, and identifying a change in at least one class of said one or more seizure classes comprises identifying a difference between a seizure class identified in the first classification analysis and a seizure class identified in the second classification analysis.

4. The method of claim 3 wherein classes are identified by at least one mathematical analysis operation selected from a statistical analysis, a graphical analysis, an unsupervised machine learning analysis, a supervised machine learning analysis, and a semi-supervised machine learning analysis.

5. The method of claim 4 wherein the statistical analysis comprises one or more of identifying a measure of central tendency of the class based on the at least one seizure metric, determining one or more percentile values based on the at least one seizure metric, determining one or more distributions based on the at least one seizure metric.

6. The method of claim 4, further comprising the step of receiving training set data for a plurality of seizures comprising a seizure class wherein the unsupervised machine learning analysis comprises a clustering analysis selected from a categorical mixture modeling analysis, a K-means clustering analysis, an agglomerative hierarchical clustering analysis, a divisive hierarchical clustering analysis, a principal component analysis, a regression algorithm, an independent component analysis, a categorical sequence labeling algorithm, and an unsupervised Hidden Markov Model sequence.

7. The method of claim 4, wherein classes are identified by more than one mathematical analysis operation.

8. The method of claim 1, wherein determining at least one seizure metric comprises determining at least two seizure metric values;

wherein performing a first classification analysis comprises identifying one or more seizure classes by determining one or more relationships among at least a portion of the plurality of seizure events, wherein the one or more relationships are based on the at least two seizure metric values for each seizure event;

wherein performing a second classification analysis comprises identifying one or more seizure classes by determining one or more relationships among at least a portion of the plurality of seizure events and the at least one additional seizure event, wherein the one or more relationships are based on the at least two seizure metric values for each seizure event; and

wherein comparing the results of the first classification analysis and the second classification analysis comprises identifying a change in at least one seizure class of said one or more seizure classes from said first classification analysis to said second classification analysis.

9. The method of claim 2, wherein the change in at least one class in moving from said first classification analysis to said second classification analysis is at least one of:

a shift in a centroid of said class;

a change in area defined by said class;

a change in a shape defined by said class;

a change in density characterizing said class.

10. The method of claim 2, further comprising identifying a change in a relationship between a first class and a second class of seizures in moving from the first classification analysis to the second classification analysis.

11. The method of claim 10, wherein the change in the relationship comprises at least one of:

a change in the distance between centroids of the first and second classes; and

a change in distance between the two closest points of the first and second classes.

12. The method of claim 1, wherein the at least one seizure metric value comprises at least one of:

a seizure severity index,

an inter-seizure interval,

a seizure frequency per unit time,

a seizure duration,

a post-ictal energy level,

a patient posture at the time of the seizure,

a patient wake state at the time of the seizure,

a time of day at which the seizure occurs,

a level of responsiveness associated with the seizure,

a level of cognitive awareness associated with the seizure,

a patient fitness level at the time of the seizure,

a patient seizure impact, and

a rate of change of one of the foregoing over at least one of a microscopic, mesoscopic or macroscopic time scale.

13. A method comprising:

identifying at least three initial seizure events in a patient;

classifying each initial seizure event into at least a first class;

identifying at least one additional seizure event;

re-classifying the first class based upon at least one of the initial seizure events and the at least one additional seizure event; and

performing a responsive action based upon the re-classifying.

14. A non-transitory computer readable program storage unit encoded with instructions that, when executed by a computer, perform a method comprising:

detecting a plurality of seizure events based upon body data of the patient;

determining, for each seizure event, at least one seizure metric value characterizing the seizure event, wherein each of said at least one seizure metric values comprises one of an autonomic index, a neurologic index, a metabolic index, an endocrine index, a tissue index, or a tissue stress index;

performing a first classification analysis of a first portion of the plurality of seizure events, the classification analysis comprising assigning each seizure event in the first portion to at least one seizure class based upon the proximity of the seizure metric values to each other;

performing a second classification analysis of a second portion of the plurality of seizure events, the classification analysis comprising assigning each seizure event in the second portion to at least one seizure class based upon the proximity of the seizure metric values, wherein said second portion comprises at least one seizure event not present in the first portion;

comparing the results of the first classification analysis and the second classification analysis; and

performing a further action selected from;

a. reporting a change from the first classification to the second classification;

b. reporting the absence of a change from the first classification to the second classification;

c. displaying a result of at least one of the first classification analysis, the second classification analysis, and the comparing;

d. identifying the emergence of a new class based on the comparing;

e. identifying the disappearance of a prior class based on the comparing;

f. identifying one or more outlier seizure events not part of any class;

g. identifying an effect of a therapy;

h. providing a therapy to the patient in response to the comparing;

i. identifying a proposed change in therapy;

j. identifying a proposed additional therapy;

k. identifying an extreme seizure event.

l. identifying a worsening trend in the patient'"'"'s seizures;

m. identifying an improvement trend in the patient'"'"'s seizures;

n. downgrading the patient'"'"'s condition in response to a worsening in the patient'"'"'s seizures; and

o. upgrading the patient'"'"'s condition in response to an improvement in the patient'"'"'s seizures.

15. The non-transitory computer readable program storage unit of claim 14, wherein said performing a further action comprises displaying at least one of a graphical depiction or a numerical representation of at least one of the first classification analysis, the second classification analysis, and the comparing.

16. The non-transitory computer readable program storage unit of claim 14,

wherein performing a first classification analysis comprises identifying one or more seizure classes by determining one or more relationships among at least a portion of the plurality of seizure events, wherein the one or more relationships are based on the at least one seizure metric value for each seizure event;

wherein performing a second classification analysis comprises identifying one or more seizure classes by determining one or more relationships among at least a portion of the plurality of seizure events and the at least one additional seizure event, wherein the one or more relationships are based on the at least one seizure metric value for each seizure event; and

wherein comparing the results of the first classification analysis and the second classification analysis comprises identifying a change in at least one class of said one or more seizure classes from said first classification analysis to said second classification analysis.

17. The non-transitory computer readable program storage unit of claim 14,

wherein determining at least one seizure metric comprises determining at least two seizure metric values;

wherein performing a first classification analysis comprises identifying one or more seizure classes by determining one or more relationships among at least a portion of the plurality of seizure events, wherein the one or more relationships are based on at least two seizure metric values for each seizure event;

wherein performing a second classification analysis comprises identifying one or more seizure classes by determining one or more relationships among at least a portion of the plurality of seizure events and the at least one additional seizure event, wherein the one or more relationships are based on at least two seizure metric values for each seizure event; and

wherein comparing the results of the first classification analysis and the second classification analysis comprises identifying a change in at least one seizure class of said one or more seizure classes from said first classification analysis to said second classification analysis.

18. The non-transitory computer readable program storage unit of claim 14, further comprising identifying a change in a relationship between a first class and a second class of seizures in moving from the first classification analysis to the second classification analysis.

19. The non-transitory computer readable program storage unit of claim 14, wherein the at least one seizure metric value comprises at least one of:

a seizure severity index,

an inter-seizure interval,

a seizure frequency per unit time,

a seizure duration,

a post-ictal energy level,

a patient posture at the time of the seizure,

a patient wake state at the time of the seizure,

a time of day at which the seizure occurs,

a level of responsiveness associated with the seizure,

a level of cognitive awareness associated with the seizure,

a patient fitness level at the time of the seizure,

a patient seizure impact, and

a rate of change of one of the foregoing over at least one of a microscopic, mesoscopic or macroscopic time scale.

20. A non-transitory computer readable program storage unit encoded with instructions that, when executed by a computer, perform a method comprising:

detecting a plurality of seizure events in a first time period, wherein each of the seizure events is detected based upon body data of the patient;

determining at least one seizure metric value for each seizure event of the plurality of seizure events;

performing a first classification analysis of a first portion of the plurality of seizure events, wherein the detection of each seizure in the first portion occurred within a second time period within said first time period, wherein said first classification analysis comprises identifying at least a first seizure class and a second seizure class based on the at least one seizure metric value, wherein the second seizure class comprises seizures that are more severe than seizures in the first seizure class;

performing a second classification analysis of a second portion of the plurality of seizure events, wherein the detection of each seizure in the second portion occurred within a third time period, wherein said third time period is a period within said first time period and wherein at least a portion of said third time period is not within the second time period, wherein said second classification analysis comprises determining, for each seizure event in said third time period, whether the seizure event is within the first seizure class and within the second seizure class, based on the at least one seizure metric value;

identifying at least one of a change in the first seizure class and the second seizure class between the first classification analysis and the second classification analysis; and

performing a responsive action based on the identifying.