Prosthetic limb gains more natural control through hand–brain connection

Researchers are paving the way for the design of bionic limbs that feel natural to users. They demonstrate the connection between hand movement patterns and motoneuron control patterns. The study, published in Science Robotics, also reports the application of these findings to a soft prosthetic hand, which was successfully tested by individuals with physical impairments.

The research study sees the collaboration of two research teams, one at Istituto Italiano di Tecnologia (Italian Institute of Technology) in Genova, Italy, led by Antonio Bicchi, and Imperial College London, UK led by Dario Farina. It is the outcome of the project “Natural BionicS” whose goal is to move beyond the model of current prosthetic limbs, which are often abandoned by patients because they do not respond in a “natural” way to their movement and control needs.

For the central nervous system to recognize the bionic limb as “natural,” it is essential for the prosthesis to interact with the environment in the same way a real limb would. For this reason, researchers believe that the prostheses should be designed based on the theory of sensorimotor synergies and soft robotics technologies, first proposed by Antonio Bicchi’s group at IIT, such as the Soft-Hand robotic hand.

If a natural-feeling interface between our nervous system and an artificial body is established, implications could go even beyond prosthetics—e.g. to allow seamless integration of humans with robot parts to assist, empower, and extend ourselves.

The study shows for the first time that two fundamental structures that organize our body, i.e. synergies at the level of spinal motoneurons and those at the level of hand behaviors, are linked. Synergies are the coordinated patterns of muscle activation and joint movements of the human body.

Researchers discovered that hand postures can be interpreted as the observable outcomes of underlying neural structures within the central nervous system. These structures can be accessed and decoded using advanced algorithms applied to the electric signals produced by our muscles. These signals are the peripheral manifestation of the activity of neural cells in the spinal cord that drive muscle contractions.

Once the activity of these cells is decoded, it is possible to identify specific cell groupings that underlie the hand behavior. This breakthrough not only enhances the understanding of the neural mechanisms driving motor control but also opens new avenues for developing more intuitive and effective human-machine interfaces.

Researchers can now co-design multi-synergistic robotic hands and neural decoding algorithms, allowing prosthetic users to achieve natural control to span infinite postures and execute dexterous tasks, including in-hand manipulation, not feasible with other approaches.

More specifically, the researchers designed a soft prosthetic hand with two degrees of actuation, enabling it to perform postures driven by two primary postural synergies. This innovative design was tested in real-time scenarios with 11 participants without physical impairments and three prosthesis users.

To achieve seamless control, the team developed an advanced online method that maps decoded neural synergies into the continuous operation of the two-synergy prosthetic hand. The results demonstrated that integrating neural and postural synergies allows for precise, natural, and coordinated control of multidigit actions. This approach not only ensures smoother and more intuitive movements but also represents a significant step forward in creating prosthetic devices that closely mimic the functionality and fluidity of natural limbs.

Such advancements have profound implications for improving the quality of life for prosthesis users, offering them greater autonomy and a more natural connection to their artificial limbs.