System and method for brain state classification
View Patent ↗A method for creating brain-machine interface based on received and filtered brain signals from a subject, classifying the filtered signals to classes of brain states, detecting one brain state from the classified brain states and executing by a machine a command that corresponds to the detected state.
1. A brain-machine interface system, comprising:
a monitoring device, comprising at least one sensor adapted to detect brain signals;
a processor, in communication with the monitoring device, wherein the processor is configured to analyze the signals detected by the at least one sensor, the processor comprising:
a classification module, the classification module being configured to:
arrange the received brain signals in time-windows of symmetrically overlapping sample windows with temporal sequence latencies repeating at larger factors per window to receive a predefined number of unique windows for a second of brain signals;
filter the brain signals in the sample windows to transform the signals wavelet by convolution of the signal; and
classify brain states corresponding to the analyzed signals based on the filtered sample windows; and
a user interface, coupled to the processor, the user interface to execute commands corresponding to a deduced state from the classification module,
wherein the processor is further configured to detect brain signals corresponding to imaginary movement.
2. The system of claim 1 , wherein the monitoring device is wearable and configured to fit onto the head of a subject.
3. The system of claim 1 , comprising a plurality of sensors that are spatially distributed at the monitoring device.
4. The system of claim 1 , wherein the processor further comprises a signal acquisition module to receive signals detected by the at least one sensor and to transfer the signals to pre-processing module.
5. The system of claim 1 , further comprising at least one memory module to store at least some of the detected data.
6. The system of claim 1 , wherein the classification module comprises a multi-class support vector machine (SVM) classifier.
7. The system of claim 1 , wherein the communication between the processor and the monitoring device is wireless.