Action recognition using bio electrical signals

Project Title

Action recognition using bio electrical signals

Project Area of Specialization NeuroTechProject Summary

Like the activity signals generated by our heart, which are called ECG signals, similarly our brain also generates its own activity signals Electroencephalogram (EEG), each time it processes something. These activity signals are captured in the form of micro-volt level signals. Each time we do something or we ‘think about doing something’, our brain generates distinct signals. Our aim is to detect and quantify features of brain signals that indicate the user's intentions and to translate these features in real time into device commands that accomplish certain operations. In our proposed model these signals will be acquired with a headset of electrodes and then these acquired signals will be processed. After processing these signals, they will be classified on basis of their respective cognitive actions. These classified signals will be fed to a controller for further control of certain operations. This mechanism can be used to control a complete electrical or electromechanical device such as, patients bed, wheelchair, car, all just by means of brain cognitive actions and not using arms or limbs physically. Thus, our project will help severally disabled people who cannot even use voice-controlled devices. Our mechanism will help such patients to interact and control operations conveniently. Moreover, this model will also help normal individuals as they will be able to control various operations just by sitting in their beds and only using their brains. Last it can be used for brain health monitoring or mental state and fatigue. Finally, the important aspect of the project is to make an affordable solution model and our project is using brain signals associated with motor responses of GCS 6 relayed to somatic nervous system for muscle movement.

Project Objectives

To design and develop an algorithm to classify different human actions.

Project Implementation Method

For the Signal Acquisition of our project, we are using multiple-electrode headsets such as the Emotiv EPOC (Emotiv, 2013) with 14 channels on it. Emotiv EEG neuroheadset was chosen as the signal acquisition device because its cost is not too high, it has multiple electrodes (14 channel) and it provides Software development kit (SDK) for developing BCI application on PC. This neuroheadset has 14 electrodes located over 10-20 international system positions (10-20 system (EEG) Wikipedia, 2013) AF3, F7, F3, FC5, T7, P7, O1, O2, P8, T8, FC6, F4, F8, and AF4.

For Signal Processor and Device Receiver, we use our laptop as signal processor and device receiver. Emotiv EEG neuroheadset provides wireless USB that can plug into PC for receiving EEG data from the neuroheadset. In following section, it describes how to develop BCI application on PC with Emotiv EEG device.

We wet Emotiv EEG neuroheadset foam-tipped sensors with saline solution before using it then we inserted those foam-tipped sensors into the black plastic headset arms.

After the sensor placement in the headset’s arms, we pair the EMOTIV EPOC headset to the USB receiver. We follow the 10-20 system of electrode placement to wear it on our heads, in order to maintain accuracy and receiving right kind of signals from the right area of brain.

We recorded data on the testbench. And extracted its features using various algorithms.

Benefits of the Project

This algorithm can be used to control a complete electrical or electromechanical device such as, patient’s bed, wheelchair, car, etc. All just by means of brain cognitive actions and not using arms or limbs physically. Thus, our project will help severally disabled people who cannot even use voice-controlled devices. Our mechanism will help such patients to interact and control operations conveniently. Moreover, this model will also help normal individuals as they will be able to control various operations just by sitting in their beds and only using their brains. Last it can be used for brain health monitoring or mental state and fatigue.

Technical Details of Final Deliverable

To design and develop an algorithm to classify different human actions.

Final Deliverable of the Project Software SystemType of Industry Medical Technologies NeuroTech, Wearables and ImplantablesSustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	70000
Emotiv, 2013	Equipment	1	70000	70000