5.2 Input and output modalities in BMIs

Brain-Machine Interfaces (BMIs) are revolutionizing how we interact with technology using our thoughts. These systems use different input methods to capture brain signals, from non-invasive EEG to highly invasive intracortical recordings, each with its own pros and cons.

BMIs can control various outputs, like robotic arms or computer interfaces, to help people with disabilities. However, challenges remain in signal processing, providing feedback, and ensuring long-term stability and user adaptation. Overcoming these hurdles is key to making BMIs more practical and accessible.

Input Modalities in BMI Systems

Input modalities for BMI systems

Top images from around the web for Input modalities for BMI systems

• 

  Frontiers | Systemic Review on Transcranial Electrical Stimulation Parameters and EEG/fNIRS ... View original

  Is this image relevant?

• 

  Frontiers | Design and Optimization of an EEG-Based Brain Machine Interface (BMI) to an Upper ... View original

  Is this image relevant?

• 

  Frontiers | ECoG Beta Suppression and Modulation During Finger Extension and Flexion | Neuroscience View original

  Is this image relevant?

• 

  Frontiers | Systemic Review on Transcranial Electrical Stimulation Parameters and EEG/fNIRS ... View original

  Is this image relevant?

• 

  Frontiers | Design and Optimization of an EEG-Based Brain Machine Interface (BMI) to an Upper ... View original

  Is this image relevant?

1 of 3

Top images from around the web for Input modalities for BMI systems

• 

  Frontiers | Systemic Review on Transcranial Electrical Stimulation Parameters and EEG/fNIRS ... View original

  Is this image relevant?

• 

  Frontiers | Design and Optimization of an EEG-Based Brain Machine Interface (BMI) to an Upper ... View original

  Is this image relevant?

• 

  Frontiers | ECoG Beta Suppression and Modulation During Finger Extension and Flexion | Neuroscience View original

  Is this image relevant?

• 

  Frontiers | Systemic Review on Transcranial Electrical Stimulation Parameters and EEG/fNIRS ... View original

  Is this image relevant?

• 

  Frontiers | Design and Optimization of an EEG-Based Brain Machine Interface (BMI) to an Upper ... View original

  Is this image relevant?

1 of 3

• Electroencephalography (EEG) records electrical activity non-invasively from the scalp, measuring voltage fluctuations resulting from ionic current flows within neurons
• Electrocorticography (ECoG) records electrical activity invasively directly from the surface of the brain, requiring surgical placement of electrodes on the exposed cortical surface
• Intracortical recordings are highly invasive, recording electrical activity from individual neurons or small populations of neurons using microelectrode arrays implanted directly into the cortex

Comparison of BMI input modalities

• EEG advantages: non-invasive, relatively inexpensive, minimal risk to the user, suitable for long-term use
  • Disadvantages: low spatial resolution due to signal attenuation by the skull and scalp, susceptible to artifacts from muscle activity (electromyography) and eye movements (electrooculography)
• ECoG advantages: higher spatial resolution compared to EEG, less susceptible to artifacts, provides more detailed information about localized brain activity
  • Disadvantages: requires invasive surgery to implant electrodes, risk of infection and other complications, limited to recording from the brain surface
• Intracortical recordings advantages: highest spatial and temporal resolution, allows recording of individual neuron activity, enables more precise control of BMI devices
  • Disadvantages: highly invasive requiring microelectrode array implantation into the brain, increased risk of tissue damage and immune response, limited longevity of implanted electrodes due to tissue scarring and signal degradation

Output Modalities in BMI Systems

Output modalities in BMI systems

• Robotic actuators are mechanical devices that convert electrical signals into physical motion, used to control robotic arms, hands, or other effectors, enabling the user to interact with the environment through the BMI system
• Prosthetic devices are artificial devices designed to replace missing limbs (robotic prosthetic arms, legs, hands) or enhance existing limb function, controlled by the BMI system to restore motor function or provide sensory feedback
• Computer interfaces enable the user to interact with computers or other digital devices using the BMI system, allowing for the control of cursors, virtual keyboards, or other software applications, providing a means for communication, environmental control (smart home devices), or entertainment (video games)

Integration challenges for BMI performance

• Signal processing and feature extraction involve developing algorithms to accurately decode neural signals and map them to the desired output commands while dealing with the variability and non-stationarity of neural signals across users and over time
• Feedback and closed-loop control challenges include:
  1. Providing meaningful sensory feedback (tactile, proprioceptive) to the user to enhance BMI performance and user experience
  2. Implementing closed-loop control systems that adapt to the user's intentions and optimize the BMI's performance
• Biocompatibility and long-term stability issues involve ensuring the long-term functionality and safety of implanted electrodes and devices while addressing tissue scarring, signal degradation, and potential infections
• User training and adaptation challenges include:
  1. Developing effective training protocols to help users learn to control the BMI system
  2. Accommodating individual differences in neural activity and learning rates
  3. Promoting neural plasticity to enhance BMI performance over time