Title : The application of brain-machine interface for prosthetic limb development
Abstract:
The applications of Brain-Machine Interface (BMI) are relatively recent. It is believed that BMI first started being researched and tested in the 1970’s. In its first applications, noninvasive methods were used to control a cursor-like graphical object on a machine screen through the transmission of electrical signals that were collected from the brain. Currently the applications of BMI are mainly researched in the medical field for development of robotic prosthetic limbs, awareness detection for people in a coma or vegetative state, and recovery for patients who suffered a stroke.
The use of BMI in neural prosthetics and bionic limbs is based on detecting and quantifying brain signals from the patient. After this the BMI translates the patient’s intent to device commands, making the prosthetic limb function as if it is a part of the patient’s body. This system consists of four main steps: signal acquisition, feature extraction, feature translation, and device output. Signal acquisition is generally collected through non-invasive methods such as electroencephalograms (EEG’s) or invasive methods like electrocorticography (ECoG). To provide the most productive signal acquisition, the limb being replaced and the functions of that limb are both considered in order to choose the most optimal placement of the device (e.g. EEG or ECoG). Feature extraction, the next step, is the process by which the signal collected from the patient is analyzed to determine whether the signal can be used for the prosthetic limb or if it is a sort of extraneous content. Through feature translation the collected signal features are passed onto a translational algorithm where the brain signals are converted into commands for the prosthetic limb. Lastly these commands are applied in the prosthetic limb resulting in the desired movement. In order to translate the signals collected from the brain, BMI systems rely on machine learning. Therefore, development in machine learning and artificial intelligence will also benefit the quality of BMI systems.
Neural prosthetics go beyond device output, they also aim to send signals collected from the prosthetic limb to the brain. Through this, the patient not only can regain motor skills but also regain a certain level of sensation. The restoration of natural sensory feedback from the prosthetic limb to the patient is still being researched and developed. If this restoration can be provided the amputee could live a safer and healthier lifestyle. Therefore, further research and development of BMI systems can increase life quality, recovery rates, and further research in neuroscience.
Audience Take Away
- The audience will have a general understanding of how a brain-machine interface works and the fields in which it is used
- The audience will learn about both the benefits and limitations of brain-machine interface systems
- The audience will learn about how this system is used to develop more efficient and optimal robotic prosthetic limbs
- The audience will understand the role of machine learning and artificial intelligence in the development of brain machine interface systems
- Through this presentation the audience will have a better understanding of how machine learning, artificial intelligence, and brain-machine interface systems can be developed in new fields of research, allowing for new developments in the field of neuroscience
