# US 2010/0168604 A1 · Processing for Multi-Channel Signals

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2010/0168604 A1
- Document ID: US20100168604A1
- Status: Published application
- Kind: A1
- Application number: 12/649,098
- Filed: 2009-12-29
- Published: 2010-07-01
- Earliest claimed: 2008-12-29
- Assignees: Cyberonics, Inc.
- Inventors: Javier Ramon Echauz; David E. Snyder; Kent W. Leyde
- Web page: https://patentgrep.com/patent/US20100168604A1

## CPC

- A61B 5/72: {Signal processing specially adapted for physiological signals or for diagnostic purposes}
- A61B 5/37: Intracranial electroencephalography [IC-EEG], e.g. electrocorticography [ECoG]
- A61B 5/4094: {Diagnosing or monitoring seizure diseases, e.g. epilepsy}
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)

## Abstract

Method and apparatus for improved processing for multi-channel signals. In an exemplary embodiment, an anomaly metric is computed for a multi-channel signal over a time window. The magnitude of the anomaly metric may be used to determine whether an anomaly is present in the multi-channel signal over the time window. In an exemplary embodiment, the anomaly metric may be a condition number associated with the singular values of the multi-channel signal over the time window, as further adjusted by the number of channels to produce a data condition number. Applications of the anomaly metric computation include the scrubbing of signal archives for epileptic seizure detection/prediction/counter-prediction algorithm training, pre-processing of multi-channel signals for real-time monitoring of bio-systems, and boot-up and/or adaptive self-checking of such systems during normal operation.

## Claims

### Claim 1

A method for detecting anomalies in a multi-channel signal, the method comprising:
sampling the multi-channel signal over a time window;
computing an anomaly metric for the multi-channel signal over the time window; and
identifying the presence of an anomaly based on the magnitude of the anomaly metric.

### Claim 2 (depends on 1)

The method of claim 1, the computing an anomaly metric comprising:
computing a condition number of the multi-channel signal over the time window;
adjusting the condition number based on a parameter of the multi-channel signal to generate a data condition number (DCN); and the identifying the presence of an anomaly comprising comparing the magnitude of the data condition number (DCN) to at least one threshold.

### Claim 3 (depends on 2)

The method of claim 2, further comprising:
sampling the outputs of a plurality of electrodes implanted on or in a patient to derive the multi-channel signal.

### Claim 4 (depends on 3)

The method of claim 3, the plurality of electrodes implanted on or in the patient's brain.

### Claim 5 (depends on 3)

The method of claim 3, further comprising:
wirelessly receiving the sampled outputs at a wireless receiver unit.

### Claim 6 (depends on 5)

The method of claim 5, further comprising:
identifying the anomaly during a boot-up phase of a computation device used to process the received sampled outputs.

### Claim 7 (depends on 1)

The method of claim 1, further comprising:
updating an anomaly data service over time, the anomaly data service indicating at least one threshold for comparing the magnitude of the anomaly metric to identify the presence of an anomaly; the identifying the presence of an anomaly comprising comparing the anomaly metric to the at least one threshold indicated by the anomaly data service.

### Claim 8 (depends on 7)

The method of claim 7, further comprising:
transmitting information wirelessly from the updated anomaly data service to a computation device used to process the received sampled outputs.

### Claim 9 (depends on 7)

The method of claim 7, the updating the anomaly data service over time comprising:
sampling multi-channel signals from a plurality of seizure detection systems; and
generating the at least one threshold indicated by the anomaly data service based on the sampling the multi-channel signals.

### Claim 10 (depends on 4)

The method of claim 4, further comprising:
processing the received sampled outputs using a seizure detection algorithm;
based on the identified presence of an anomaly, modifying the processing of the received sampled outputs using the seizure detection algorithm.

### Claim 11 (depends on 10)

The method of claim 10, the modifying the processing comprising omitting from the processing portions of the sampled outputs corresponding to identified anomalies.

### Claim 12 (depends on 11)

The method of claim 11, the processing comprising adjusting parameters of an adaptive seizure detection algorithm, the method further comprising detecting a seizure based on the received sampled outputs.

### Claim 13 (depends on 4)

The method of claim 4, further comprising:
processing the received sampled outputs using a seizure prediction algorithm; and
based on the identified presence of an anomaly, modifying the processing of the received sampled outputs using the seizure prediction algorithm.

### Claim 14 (depends on 4)

The method of claim 4, further comprising:
processing the received sampled outputs using a seizure likelihood computation algorithm; and
based on the identified presence of an anomaly, modifying the processing of the received sampled outputs using the seizure likelihood computation algorithm.

### Claim 15 (depends on 2)

The method of claim 2, the multi-channel signal comprising a signal sampled from a plurality of electrodes implanted on or in a patient's brain, the method further comprising:
generating a DCN time series corresponding to a plurality of time windows;
generating an anomaly log based on the DCN time series;
merging anomalies in the anomaly log separated by less than a minimum separation to generate a modified anomaly log;
identifying segments of the multi-channel signal corresponding to anomalies in the modified anomaly log; and
outputting time-expanded versions of the identified segments to a record.

### Claim 16 (depends on 2)

The method of claim 2, the computing the condition number comprising:
generating a matrix A[k] for a time window k, each column of the matrix A[k] comprising sampled values of a single channel of the multi-channels signal over a time window k;
computing a maximum singular value of the matrix A[k];
computing a minimum singular value of the matrix A[k]; and
dividing the maximum singular value by the minimum singular value to generate the condition number.

### Claim 17 (depends on 2)

The method of claim 2, the computing the condition number comprising:
generating a square matrix A[k] for a time window k, each column of the matrix A[k] comprising sampled values of a single channel of the multi-channels signal over a time window k;
deriving a norm of the matrix A[k];
deriving the norm of the matrix A−1[k]; and
dividing the norm of the matrix A[k] by the norm of the matrix A−1[k] to generate the condition number.

### Claim 18 (depends on 17)

The method of claim 17, the norm being an L1-norm.

### Claim 19 (depends on 17)

The method of claim 17, the norm being an L-infinity norm.

### Claim 20 (depends on 2)

The method of claim 2, the adjusting the condition number based on a parameter of the multi-channel signal to generate a data condition number (DCN) comprising computing:
D


C


N
=
1
+
(
C
-
1
)
N
,
wherein C is the condition number, and N is the number of channels in the multi-channel signal.

### Claim 21 (depends on 2)

The method of claim 2, the parameters used to adjust the condition number comprising at least one of window size, signal measurement bandwidth, and electrode montage.

### Claim 22

An apparatus for processing a multi-channel signal, the apparatus comprising:
an anomaly metric computation module configured to compute an anomaly metric for the multi-channel signal over a time window; and
an anomaly identification module configured to identify the presence of an anomaly in the multi-channel signal based on the magnitude of the anomaly metric.

### Claim 23 (depends on 22)

The apparatus of claim 22, the anomaly metric computation module configured to compute the anomaly metric by computing a condition number of the multi-channel signal over the time window; the anomaly metric computation module further configured to adjust the condition number by a parameter of the multi-channel signal to generate a data condition number (DCN); the anomaly identification module configured to identify the presence of the an anomaly by comparing the magnitude of the DCN to at least one threshold.

### Claim 24 (depends on 23)

The apparatus of claim 23, the apparatus further comprising a plurality of electrodes configured to be implanted on or in a patient to derive the multi-channel signal.

### Claim 25 (depends on 24)

The apparatus of claim 24, the plurality of electrodes implanted on or in the patient's brain.

### Claim 26 (depends on 24)

The apparatus of claim 24, the anomaly metric computation module and anomaly identification module residing in a wireless receiver unit, the wireless receiver unit wirelessly receiving the multi-channel signal from the plurality of electrodes.

### Claim 27 (depends on 26)

The apparatus of claim 26, the anomaly identification module further configured to identify an anomaly during a boot-up phase of a computation device used to process the received sampled outputs.

### Claim 28 (depends on 22)

The apparatus of claim 22, further comprising:
an anomaly data service configured to be updated over time, the anomaly data service configured to indicate at least one threshold for comparing the magnitude of the anomaly metric to identify the presence of an anomaly; the anomaly identification module configured to compare the magnitude of the anomaly metric to the at least one threshold indicated by the anomaly data service.

### Claim 29 (depends on 28)

The apparatus of claim 28, further comprising a wireless receiver for receiving information from the updated anomaly data service.

### Claim 30 (depends on 25)

The apparatus of claim 25, the apparatus used to detect seizures in a patient.

### Claim 31 (depends on 25)

The apparatus of claim 25, the apparatus used to predict seizures in a patient.

### Claim 32 (depends on 25)

The apparatus of claim 25, the apparatus used to compute the likelihood of a patient entering into a seizure.

### Claim 33 (depends on 23)

The apparatus of claim 23, the multi-channel signal comprising multi-channel signal comprising a signal sampled from a plurality of electrodes implanted on or in a patient's brain, the apparatus further configured to:
generate a DCN time series corresponding to a plurality of time windows;
generate an anomaly log based on the DCN time series;
merge anomalies in the anomaly log separated by less than a minimum separation to generate a modified anomaly log;
identify segments of the multi-channel signal corresponding to anomalies in the modified anomaly log; and
output time-expanded versions of the identified segments to a record.

### Claim 34 (depends on 23)

The apparatus of claim 23, the anomaly metric computation module configured to:
generate a matrix A[k] for a time window k, each column of the matrix A[k] comprising sampled values of a single channel of the multi-channels signal over a time window k;
compute a maximum singular value of the matrix A[k];
compute a minimum singular value of the matrix A[k]; and
divide the maximum singular value by the minimum singular value to generate the condition number.

### Claim 35 (depends on 23)

The apparatus of claim 23, the anomaly metric computation module configured to:
generate a square matrix A[k] for a time window k, each column of the matrix A[k] comprising sampled values of a single channel of the multi-channels signal over a time window k;
derive a norm of the matrix A[k];
derive the norm of the matrix A′−1[k]; and
divide the norm of the matrix A[k] by the norm of the matrix A−1[k] to generate the condition number.

### Claim 36 (depends on 23)

The apparatus of claim 23, the anomaly metric computation module configured to compute:
D


C


N
=
1
+
(
C
-
1
)
N
,
as the data condition number (DCN), wherein C is the condition number, and N is the number of channels in the multi-channel signal.

### Claim 37

An apparatus for processing a multi-channel signal, the apparatus comprising:
means for sampling the multi-channel signal over a time window;
means for computing an anomaly metric for the multi-channel signal over the time window; and
means for identifying the presence of an anomaly based on the magnitude of the anomaly metric.

### Claim 38

A computer program product for detecting anomalies in a multi-channel signal, the product comprising:
computer-readable medium comprising:
code for causing a computer to compute an anomaly metric for the multi-channel signal over the time window; and
code for causing a computer to identify the presence of an anomaly based on the magnitude of the anomaly metric.