EEG based communication system in generalized & customized modes for differently abled communities / Paulraj M. P. ...[et al.]

Differently abled people such as patients with Amyotrophic Lateral Sclerosis, brain stem stroke and spinal cord injury, encounter difficulty in communication due to the loss of muscle control and speech. Intelligent Brain Machine interfaces are devices which can be used to aid these severely affected people through the power of thought. In this research work, a Thought Controlled Communication System has been developed using seven English words which is considered to convey the basic needs of a patient. The proposed communication system records the Electroencephalography signal while mentally reading the words. The recorded EEG signals are pre-processed and segmented into four frequency bands. The band frequency signals are used to extract features using band power and power spectral density algorithms. In this analysis, two simple classifiers namely Multi Layer Neural Network and k-Nearest Neighbor have been used for recognizing the extracted features in both generalized and customized modes. The proposed classification system has been validated through simulation.