Neuroengineering Lab

1- Developing Closed loop Neural Interfaces for the spinal cord injury treatment

In this project, we design a closed-loop rehabilitative setup with potential for near-term use in the treatment of spinal cord patients. The objective of this project is to design an implantable neural interface on rodents with long-lasting decoding capabilities and low computational complexity. The project is conducted in collaboration with Intelligent Sensing Laboratory from Newcastle university, Neuroengineering research laboratory form  Iran science and technology university and Cognitive Neuroscience Research Center from  Kerman University of Medical Sciences. We have currently designed and fabricated 8-Channel cortical data Acquisition, 3-Channel programmable microstimulator and 8-ch microelectrode arrays.  Some sample publications:

2- Design of closed-loop cortical pattern controller in the behaving rodents

In this project, we offer a wireless and closed-loop neural pattern control based on the local field potential (LFP) signals. The objective of this project is to develop a closed-loop multichannel cortical signal processor/ stimulator that can control the cortical activity of the neural system. This  project is conducted in collaboration with Bahonar university of Kerman and Cognitive Neuroscience Research Center from  Kerman University of Medical Sciences.We have currently designed and fabricated an automated lever pressing setup that let us to measure animal forelimb extension/flexion functions with different level of difficulty on freely moving rodents.

3- Developing Advanced Cortical Signal Processing Algorithms

We are developing new signal processing algorithms to decode motor commands that introduce long-term stability with real-world performance. We are developing this machine learning strategies in different level of processing. First, in the pre-processing step novel artifact removal techniques are developed to remove the effects of different type of physiological and non-physiological artifact sources. Second, we introduce new multi-dimensional decoding algorithms to produce accurate motor commands.  Third, we incorporate adaptive algorithms into kinematic/kinetic decoder to minimize the day-to-day variable performance of brain-machine interfaces.  Some sample publications: