Spatiotemporal pattern recognition for neurological event detection and prediction in an implantable device
View Patent ↗A system and method for detecting and predicting neurological events with an implantable device uses a relatively low-power central processing unit in connection with signal processing circuitry to identify features (including half waves) and calculate window-based characteristics (including line lengths and areas under the curve of the waveform) in one or more electrographic signals received from a patient's brain. The features and window-based characteristics are employed within the framework of a programmable finite state machine to identify patterns and sequences in and across the electrographic signals, facilitating early and reliable detection and prediction of complex spatiotemporal neurological events in real time, and enabling responsive action by the implantable device.
1. A method for detecting a neurological condition, the method comprising:
receiving at least one electrical signal with at least one electrode, where the electrode is in electrical communication with neural tissue;
processing the at least one electrical signal with a detector; and
detecting the neurological condition with the detector, where the neurological condition is determined by a sequence of detections of individual neurological events, and where the detections occur in a predetermined order.
2. The method for detecting a neurological condition of claim 1 , wherein the sequence comprises a plurality of waveform segments.
3. The method for detecting a neurological condition of claim 2 , wherein the plurality of waveform segments includes a first segment and a second segment.
4. The method for detecting a neurological condition of claim 3 , wherein the first segment is observed at a first time in a first location in the patient's brain.
5. The method for detecting a neurological condition of claim 4 , wherein the second segment is observed at a second time in a second location in the patient's brain.
6. The method for detecting a neurological condition of claim 5 , wherein the second time follows the first time.
7. The method for detecting a neurological condition of claim 5 , wherein the first location is substantially the same as the second location.
8. The method for detecting a neurological condition of claim 5 , wherein the first location is substantially separate from the second location.
9. The method for detecting a neurological condition of claim 1 , wherein the sequence comprises a sequence of signal characteristics.
10. The method for detecting a neurological condition of claim 9 , wherein the sequence of signal characteristics is encoded in a finite state machine.
11. The method for detecting a neurological condition of claim 10 , wherein the finite state machine comprises a master state machine and at least one instance.
12. The method for detecting a neurological condition of claim 11 , wherein the at least one instance comprises a current state variable and an overall life remaining variable.
13. The method for detecting a neurological condition of claim 12 , wherein the at least one instance further comprises a state life remaining variable.
14. The method for detecting a neurological condition of claim 11 , wherein the identifying step comprises the steps of: setting an instance of the finite state machine to a first state; detecting a waveform characteristic in the sequence of signal characteristics; transitioning the instance to a destination state.
15. The method for detecting a neurological condition of claim 14 , wherein the instance includes a current state, and wherein the transitioning step comprises updating the current state from the first state to the destination state.
16. The method for detecting a neurological condition of claim 14 , wherein the instance includes a current state, and wherein the transitioning step comprises the steps of: determining whether a transition associated with the transitioning step is exclusive or nonexclusive; if the transition is exclusive, updating the current state from the first state to the destination state; and if the transition is nonexclusive, spawning a new instance in the destination state.
17. The method for detecting a neurological condition of claim 10 , wherein the identifying step comprises the steps of: setting the finite state machine to an initial state; detecting a first waveform characteristic in the sequence of signal characteristics; transitioning the finite state machine to a first state; detecting a second waveform characteristic in the sequence of signal characteristics; transitioning the finite state machine to a detection state; and performing an action defined in connection with the detection state.
18. The method of claim 1 further comprising treating the neurological condition in response to its detection.