Refined Force Estimation in Monkey's Pinching Tasks Through Integrated EMG and ECoG Data: A Kalman Filter Method.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Rehabilitation Science

Background:

• Accurate decoding of motor intentions is vital for effective brain-computer interfaces (BCIs).
• Current BCIs often struggle with precise force estimation, limiting their functional application.
• Electrocorticography (ECoG) offers high-resolution neural signals, but integrating peripheral information can enhance decoding.

Purpose of the Study:

• To develop an enhanced Kalman filter for improved force estimation in pinching tasks using BCIs.
• To investigate the benefit of integrating electromyography (EMG) with ECoG signals for motor output decoding.
• To enhance the accuracy of force prediction by incorporating musculoskeletal dynamics.

Main Methods:

• An enhanced Kalman filter was designed to integrate ECoG and EMG data.
• EMG signals were incorporated as a state variable within the Kalman filter framework.
• The approach was tested for force estimation accuracy in human participants performing pinching tasks.

Main Results:

• The integrated ECoG-EMG approach significantly improved force estimation accuracy compared to ECoG alone.
• Enhanced decoding performance was particularly notable during dynamic force changes.
• The inclusion of EMG data effectively captured musculoskeletal dynamics relevant to motor output.

Conclusions:

• Integrating musculoskeletal dynamics via EMG into ECoG-based BCIs substantially improves motor decoding.
• This enhanced approach holds promise for advancing prosthetic limb control.
• The findings suggest a pathway for more effective motor rehabilitation strategies for individuals with motor impairments.