Surgery – Diagnostic testing – Detecting brain electric signal
Reexamination Certificate
2001-05-18
2003-12-02
Winakur, Eric F. (Department: 3736)
Surgery
Diagnostic testing
Detecting brain electric signal
Reexamination Certificate
active
06658287
ABSTRACT:
FIELD OF THE INVENTION
The present invention is directed to predicting the onset of epileptic seizures, and more specifically to a method and apparatus for automatically interpreting information representing the activity of the brain so as to predict the onset of a seizure in order to alert a patient of the possibility of an impending seizure and/or to take preventative actions to avert a seizure.
BACKGROUND OF THE INVENTION
Epilepsy affects approximately 1% of the population in the United States and approximately 2% of the population worldwide. Of those affected by the disease, approximately one-third have seizures that cannot be controlled by medication or cured by surgery. Epilepsy surgery requires locating the region of the brain where seizure onset occurs and the pathways through which the seizures spread, a process that is not completely accurate and reliable. Moreover, epilepsy surgery is accompanied by the inherent risk of neurologic injury, disfigurement and other complications. Some individuals have epileptic seizures that cannot be controlled by standard medication, are inoperable because seizure onset is not localized, or originate from vital areas of the brain which cannot be surgically removed. These individuals may resort to high doses of intoxicating medications and/or other experimental therapies.
Several prior art algorithms for seizure prediction and/or detection are known. See, for example, U.S. Pat. No. 5,857,978, to Hively et al., entitled “Epileptic Seizure Prediction by Nonlinear Methods,” U.S. Pat. No. 3,863,625, to Viglione et al., entitled “Epileptic Seizure Warning System,” U.S. Pat. No. 4,566,464, entitled “Implantable Epilepsy Monitor Apparatus.”
It is desirable to provide a method and apparatus for predicting seizures with such accuracy that the activity of the brain can be monitored by an implantable device to warn a patient of the likelihood of an impending seizure, and/or to take preventative actions through application of intervention measures to abort or modulate the seizure prior to clinical onset.
SUMMARY OF THE INVENTION
Briefly, the present invention is directed to a method and apparatus for predicting the onset of a seizure in an individual. Whereas prior art systems and algorithms determine that a seizure is occurring after detection of its actual electrical onset, which may or may not occur before detectable clinical manifestations of a seizure, the present invention is directed to a method and apparatus for predicting that a seizure is going to occur sometime well in advance of any detectable electrical onset or clinical onset of seizure activity. The prediction achieved according to the present invention is well in advance of any electrical onset of seizures, or clinical onset, and before there are visually obvious changes in EEG patterns.
The method and apparatus according to the present invention operate by monitoring signals representing the activity of the brain, extracting features from the signals and deriving a feature vector representing a combination of those features that are determined (during “off-line” analysis of a particular individual and/or other knowledge of seizure prediction across a number of individuals) to be predictive of seizure onset, and analyzing the feature vector with a trainable algorithm implemented by, for example, a wavelet neural network, to predict seizure onset. Features are extracted on both an instantaneous basis and a historical basis. Features are collected and analyzed in different time frames, such as over days, hours, minutes, and seconds.
Preferably, the system is implemented in an implantable device that an individual or physician can interface with in much the same manner as an implantable pacemaker or defibrillator. Interface to the implantable device is by way of a body-wearable or attachable patient access unit that includes a display (such as a liquid crystal display), an audible or visible alert, a vibration alert, and a user interface (such as a button keypad). The output of the implantable device may comprise a signal(s) indicating a probability of seizure occurrence within one or more specified periods of time in parallel. The patient may program the system via the patient access unit to generate certain levels of alerts based on programmable probability thresholds. Access may also take place via connection to a local or physician's office personal computer and to a central facility via the Internet. Programming can be done by the patients with their personal unit, or the physician may choose to completely control this process via periodic checks with an office unit, the patient's home PC or via the Internet, portable cellular, infra-red, microwave or other communication device.
In addition, the system may be programmed to automatically trigger preventative actions, such as the application of an electrical shock, the delivery of one or more drugs or the activation of a pacing algorithm which can be employed to abort the seizure or mitigate the severity of a seizure. Outputs from the device may be used to train the patient in a biofeedback scheme to learn to abort seizures themselves.
A distinguishing theme of the present invention is that the most accurate seizure predictor is one based on the synergy of multiple features or a single feature artificially customized from raw data, as opposed to prior art techniques that involve reliance on a single conventional feature. Another important aspect of the invention is the generation as output of one or more probability measures, each associated with a different prediction horizon, that represent the likelihood a seizure will occur during the corresponding prediction horizon.
Another aspect of the invention a method for applying intervention measures to an animal to abort or modulate a seizure comprising the step of adjusting the modality of an intervention measure and/or parameters of an intervention measure based upon a probability measure indicative of a likelihood of seizure occurrence and/or a predicted time to seizure onset.
REFERENCES:
patent: 3837331 (1974-09-01), Ross
patent: 3850161 (1974-11-01), Liss
patent: 3863625 (1975-02-01), Viglione et al.
patent: 3967616 (1976-07-01), Ross
patent: 3993046 (1976-11-01), Fernandez et al.
patent: 4566464 (1986-01-01), Piccone et al.
patent: 4702254 (1987-10-01), Zabara
patent: 4867164 (1989-09-01), Zabara
patent: 5025807 (1991-06-01), Zabara
patent: 5304206 (1994-04-01), Baker, Jr. et al.
patent: 5311876 (1994-05-01), Olsen et al.
patent: 5713923 (1998-02-01), Ward et al.
patent: 5720294 (1998-02-01), Skinner
patent: 5743860 (1998-04-01), Hively et al.
patent: 5857978 (1999-01-01), Hively et al.
patent: 5928272 (1999-07-01), Adkins et al.
patent: 5978702 (1999-11-01), Ward et al.
patent: 5995868 (1999-11-01), Dorfmeister et al.
patent: 6016449 (2000-01-01), Fischell et al.
patent: 6018682 (2000-01-01), Rise
patent: 6061593 (2000-05-01), Fischell et al.
patent: WO97/26823 (1997-07-01), None
patent: WO 00/10455 (2000-03-01), None
Osorio, I. et al., “Real-Time Automated Detection and Quantitative Analysis of Seizures and Short-Term Prediction of Clinical Onset”, Epliepsia, vol. 39, No. 6, Jun. 1998, pp. 615-627.
“Automated Seizure Prediction Paradigm”Epilepsia39:56 (1998).
Cover “The Best Two Independent Measurements are Not the Two Best”IEEE Transactions on System, Man, and Cybernetics(Jan. 1974).
Echauz et al. “Eliptic and Radial Wavelet Neural Networks”in Proc. Second World Automation Congress, Montpellier, France, May 27-30, 19965:173-179 (1996).
Echauz et al. “Neuro-Fuzzy Approaches to decision Making: A Comparative Study with an Application to check Authorization”0J. of Intelligent and Fuzzy Sys. 6:259-278 (1998).
Elger et al. “Seizure Prediction by Nonlinear Time Series Analysis of Brain Electrical Activity,”European J. of Neuroscience10:786-789 (1998).
Grassberger et al. “Characterization of Strange Attractors”Physical Rev. Letters50 (5) :346-349.
Iasemidis et al. “Chaos Theory and Epilepsy”The Neuroscienst2 (2) :118-126 (Mar. 1996).
Iasemidis et al. “Spatiotemporal El
Echauz Javier
Esteller Rosana
Litt Brian
Vachtsevanos George
Georgia Tech Research Corporation
Needle & Rosenberg P.C.
Szmal Brian
Winakur Eric F.
LandOfFree
Method and apparatus for predicting the onset of seizures... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for predicting the onset of seizures..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for predicting the onset of seizures... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3125166