Patent application number | Description | Published |
20080269835 | METHOD AND APPARATUS FOR DETECTION OF NERVOUS SYSTEM DISORDERS - Systems and methods for detecting and/or treating nervous system disorders, such as seizures, are disclosed. Certain embodiments of the invention relate generally to implantable medical devices (IMDs) adapted to detect and treat nervous system disorders in patients with an IMD. Certain embodiments of the invention include detection of seizures based upon comparisons of long-term and short-term representations of physiological signals. Other embodiments include prediction of seizure activity based upon analysis of physiological signal levels. An embodiment of the invention monitors the quality of physiological signals, and may be able to compensate for signals of low signal quality. A further embodiment of the invention includes detection of seizure activity following the delivery of therapy. | 10-30-2008 |
20100114237 | MOOD CIRCUIT MONITORING TO CONTROL THERAPY DELIVERY - Brain signals may be monitored at different locations of a mood circuit in order to determine a mood state of the patient. A relationship (e.g., a ratio) between frequency band characteristics of the monitored brain signals may be indicative of a particular mood state. In some examples, therapy parameter values that define the therapy delivered to the patient may be selected to maintain a target relationship (e.g., a target ratio) between the frequency band characteristics of the brain signals monitored within the mood circuit. In addition, in some examples, therapy delivery to the patient may be controlled based on the frequency band characteristics of brain signals sensed at different portions of the mood circuit. | 05-06-2010 |
20100280334 | PATIENT STATE DETECTION BASED ON SUPPORT VECTOR MACHINE BASED ALGORITHM - A patient state is detected with at least one classification boundary generated by a supervised machine learning technique, such as a support vector machine. In some examples, the patient state detection is used to at least one of control the delivery of therapy to a patient, to generate a patient notification, to initiate data recording, or to evaluate a patient condition. In addition, an evaluation metric can be determined based on a feature vector, which is determined based on characteristics of a patient parameter signal, and the classification boundary. Example evaluation metrics can be based on a distance between at least one feature vector and the classification boundary and/or a trajectory of a plurality of feature vectors relative to the classification boundary over time. | 11-04-2010 |
20100280335 | PATIENT STATE DETECTION BASED ON SUPERVISED MACHINE LEARNING BASED ALGORITHM - A patient state is detected with at least one classification boundary generated by a supervised machine learning technique, such as a support vector machine. In some examples, the patient state detection is used to at least one of control the delivery of therapy to a patient, to generate a patient notification, to initiate data recording, or to evaluate a patient condition. In addition, an evaluation metric can be determined based on a feature vector, which is determined based on characteristics of a patient parameter signal, and the classification boundary. Example evaluation metrics can be based on a distance between at least one feature vector and the classification boundary and/or a trajectory of a plurality of feature vectors relative to the classification boundary over time. | 11-04-2010 |
20100280574 | PATIENT STATE DETECTION BASED ON SUPPORT VECTOR MACHINE BASED ALGORITHM - A patient state is detected with at least one classification boundary generated by a supervised machine learning technique, such as a support vector machine. In some examples, the patient state detection is used to at least one of control the delivery of therapy to a patient, to generate a patient notification, to initiate data recording, or to evaluate a patient condition. In addition, an evaluation metric can be determined based on a feature vector, which is determined based on characteristics of a patient parameter signal, and the classification boundary. Example evaluation metrics can be based on a distance between at least one feature vector and the classification boundary and/or a trajectory of a plurality of feature vectors relative to the classification boundary over time. | 11-04-2010 |
20100280579 | POSTURE STATE DETECTION - A patient state is detected with at least one classification boundary generated by a supervised machine learning technique, such as a support vector machine. The patient state can be, for example, a patient posture state. In some examples, the patient state detection is used to at least one of control the delivery of therapy to a patient, to generate a patient notification, to initiate data recording, or to evaluate a patient condition. In addition, an evaluation metric can be determined based on a feature vector, which is determined based on characteristics of a patient parameter signal, and the classification boundary. Example evaluation metrics can be based on a distance between at least one feature vector and the classification boundary and/or a trajectory of a plurality of feature vectors relative to the classification boundary over time. | 11-04-2010 |
20100292753 | Method and Apparatus for Detection of Nervous System Disorders - Systems and methods for detecting and/or treating nervous system disorders, such as seizures. Certain embodiments of the invention relate generally to implantable medical devices (IMDs) adapted to detect and treat nervous system disorders in patients with an IMD. Certain embodiments of the invention include detection of seizures based upon comparisons of long-term and short-term representations of physiological signals. Other embodiments include prediction of seizure activity based upon analysis of physiological signal levels. An embodiment of the invention monitors the quality of physiological signals, and may be able to compensate for signals of low signal quality. A further embodiment of the invention includes detection of seizure activity following the delivery of therapy. | 11-18-2010 |
20110263999 | METHOD AND APPARATUS FOR DETECTION OF NERVOUS SYSTEM DISORDERS - Systems and methods for detecting and/or treating nervous system disorders, such as seizures. Certain embodiments of the invention relate generally to implantable medical devices (IMDs) adapted to detect and treat nervous system disorders in patients with an IMD. Certain embodiments of the invention include detection of seizures based upon comparisons of long-term and short-term representations of physiological signals. Further embodiments of the invention include preparing for the delivery of therapy by warming up therapy delivery components prior to the expected delivery of therapy. | 10-27-2011 |
20120116179 | TECHNIQUES FOR DATA RETENTION UPON DETECTION OF AN EVENT IN AN IMPLANTABLE MEDICAL DEVICE - Methods and apparatus for storing data records associated with a medical monitoring event in a data structure. These include initiating loop recording in an implantable medical device upon determination of a neurological event, wherein loop recording comprises storing a data record of a plurality of data records in a data structure, the plurality of data records representing information about determined neurological events. Methods and apparatus can further include determining a priority index for the plurality of data records based on severity levels of the determined neurological events and replacing older data records of the plurality of data records on the data structure with new data records according to the priority index, wherein the new data records selectively replace those data records in the data structure having the lowest associated priority index. | 05-10-2012 |
20130053722 | Method and Apparatus for Detecting a Biomarker in the Presence of Electrical Stimulation - Various embodiments concern identifying a biomarker in the presence of electrical stimulation. Various embodiments concern delivering electrical stimulation to a patient and sensing one or more signals while the electrical stimulation is being delivered, the one or more signals including data indicative of physiological activity. Various embodiments further include determining an intensity of the electrical stimulation and determining whether the data indicates the presence of a biomarker based on a variable threshold, the variable threshold being variable based on the intensity of the electrical stimulation. Various embodiments concern determining a relationship between stimulation intensity and a biomarker parameter to determine the variability of the variable threshold. | 02-28-2013 |
20130197605 | STIMULATION ELECTRODE SELECTION - Bioelectrical signals may be sensed within a brain of a patient with a plurality of sense electrode combinations. A stimulation electrode combination for delivering stimulation to the patient to manage a patient condition may be selected based on the frequency band characteristics of the sensed signals. In some examples, a stimulation electrode combination associated with the sense electrode combination that sensed a bioelectrical brain signal having a relatively highest relative beta band power level may be selected to deliver stimulation therapy to the patient. Other frequency bands characteristics may also be used to select the stimulation electrode combination. | 08-01-2013 |
20140032512 | Techniques for Data Retention upon Detection of an Event in an Implantable Medical Device - Methods and apparatus for storing data records associated with a medical monitoring event in a data structure. These include initiating loop recording in an implantable medical device upon determination of a neurological event, wherein loop recording comprises storing a data record of a plurality of data records in a data structure, the plurality of data records representing information about determined neurological events. Methods and apparatus can further include determining a priority index for the plurality of data records based on severity levels of the determined neurological events and replacing older data records of the plurality of data records on the data structure with new data records according to the priority index, wherein the new data records selectively replace those data records in the data structure having the lowest associated priority index. | 01-30-2014 |
20140135869 | STIMULATION ELECTRODE SELECTION - Bioelectrical signals may be sensed within a brain of a patient with a plurality of sense electrode combinations. A stimulation electrode combination for delivering stimulation to the patient to manage a patient condition may be selected based on the frequency band characteristics of the sensed signals. In some examples, a stimulation electrode combination associated with the sense electrode combination that sensed a bioelectrical brain signal having a relatively highest relative beta band power level may be selected to deliver stimulation therapy to the patient. Other frequency bands characteristics may also be used to select the stimulation electrode combination. | 05-15-2014 |
20140135870 | STIMULATION ELECTRODE SELECTION - Bioelectrical signals may be sensed within a brain of a patient with a plurality of sense electrode combinations. A stimulation electrode combination for delivering stimulation to the patient to manage a patient condition may be selected based on the frequency band characteristics of the sensed signals. In some examples, a stimulation electrode combination associated with the sense electrode combination that sensed a bioelectrical brain signal having a relatively highest relative beta band power level may be selected to deliver stimulation therapy to the patient. Other frequency bands characteristics may also be used to select the stimulation electrode combination. | 05-15-2014 |
20140276185 | CONTROL OF SPECTRAL AGRESSORS IN A PHYSIOLOGICAL SIGNAL MONTORING DEVICE - This disclosure describes techniques for controlling spectral aggressors in a sensing device that uses a low power sleep mode to manage the power consumed by the device. In some examples, the techniques for controlling spectral aggressors may include configuring one or more of an algorithm processing rate for a processor, a buffering rate for the processor, a sampling rate for an analog-to-digital converter, an execution unit processing rate for the processor, and an algorithm subdivision factor for the processor such that spectral interference caused by a sleep cycle rate of the processor occurs outside of one or more target frequency bands of a sampled signal. The techniques of this disclosure may be used to reduce noise in a sensing system that uses a low power sleep mode to manage the power consumed by the device. | 09-18-2014 |
20140276186 | CONTROL OF SPECTRAL AGRESSORS IN A PHYSIOLOGICAL SIGNAL MONTORING DEVICE - This disclosure describes techniques for controlling spectral aggressors in a sensing device that uses a chopper amplifier to amplify an input signal prior to sampling the signal. In some examples, the techniques for controlling spectral aggressors may include generating a chopper-stabilized amplified version of an input signal based on a chopper frequency, sampling the chopper-stabilized amplified version of the input signal at a sampling rate to generate a sampled signal, and analyzing a target frequency band of the sampled signal. The chopper frequency and the sampling rate may cause spectral interference that is generated due to the chopper frequency to occur in the sampled signal at one or more frequencies that are outside of the target frequency band of the sampled signal. The techniques for controlling spectral aggressors may reduce the noise caused by the chopper frequency in the resulting sampled signal, thereby improving the quality of the signal. | 09-18-2014 |
20140309614 | Clustering of Recorded Patient Neurological Activity to Determine Length of a Neurological Event - Apparatus and method detect a detection cluster that is associated with a neurological event, such as a seizure, of a nervous system disorder and update therapy parameters that are associated with a treatment therapy. The occurrence of the detection cluster is detected when the maximal ratio exceeds an intensity threshold. If the maximal ratio drops below the intensity threshold for a time interval that is less than a time threshold and subsequently rises above the intensity threshold, the subsequent time duration is considered as being associated with the detection cluster rather than being associated with a different detection cluster. Consequently, treatment of the nervous system disorder during the corresponding time period is in accordance with one detection cluster. Treatment therapy may be provided by providing electrical stimulation, drug infusion or a combination. Therapy parameters may be updated for each m | 10-16-2014 |
20150080674 | TECHNIQUES FOR DATA RETENTION UPON DETECTION OF AN EVENT IN AN IMPLANTABLE MEDICAL DEVICE - Methods and apparatus for storing data records associated with a medical monitoring event in a data structure. These include initiating loop recording in an implantable medical device upon determination of a neurological event, wherein loop recording comprises storing a data record of a plurality of data records in a data structure, the plurality of data records representing information about determined neurological events. Methods and apparatus can further include determining a priority index for the plurality of data records based on severity levels of the determined neurological events and replacing older data records of the plurality of data records on the data structure with new data records according to the priority index, wherein the new data records selectively replace those data records in the data structure having the lowest associated priority index. | 03-19-2015 |