8 apr. 2025
Teodora Gheorghe
AI-Driven Prosthetic Algorithms: A New Era for Amputees
In an era where technology and medicine converge to enhance the quality of life, the NerveRepack project represents a significant leap forward in the field of prosthetic limbs. This innovative initiative aims to develop a new type of arm prosthesis controlled directly by neural signals, providing users with not only mobility but also tactile sensations. By redefining the interaction between humans and prosthetic devices, NerveRepack stands at the forefront of neuroprosthetic advancements.
A New Generation of Arm Prostheses for Amputees
The NerveRepack project is dedicated to creating a revolutionary prosthesis for individuals with forearm amputations. Unlike traditional prosthetic solutions, this device will be controlled through neural signals captured from the patient’s residual limb. Moreover, it will provide the user with tactile sensations from each finger while handling objects. This breakthrough is made possible by establishing a bidirectional connection between the prosthesis and the patient’s nervous system through specialized electrodes implanted in the median and ulnar nerves.
How It Works
The innovation of the NerveRepack prosthesis lies in its ability to establish real-time communication with the nervous system through two types of implanted electrodes:
Plug electrodes: These acquire motor neural signals from the motor fascicles of the median and ulnar nerves and wirelessly transmit them to the prosthesis’ control system, enabling movement.
Feedback cuff electrodes: These receive signals from pressure sensors embedded in the prosthetic fingers and stimulate the sensory fascicles of the median and ulnar nerves to generate tactile feedback for the user.
According to the Amputee Coalition, approximately 2 million people in the United States alone live with limb loss.
The World Health Organization statistics and scientific studies revealed that in the last ten years in the world „more than 50 million people (0.5% - 0.8% of world population) had limb loss [1-2]. It is estimated that in 2018 „in US there were more than 2.2 million of amputees” and in UE countries "were more than 1.2 million of amputees” [3].
The implementation of such advanced prostheses could significantly improve the quality of life for millions worldwide, restoring both functionality and sensory perception.
Patient-Controlled Prosthesis with Motor Neural Signals
One of the core principles behind the NerveRepack project is understanding the correlation between brain commands and the motor neural signals in the median and ulnar nerves.
The Science Behind Neural Control
When a person initiates a command in the brain to move their fingers, specific trains of neural signals travel through the median and ulnar nerves to the muscles of the forearm, palm, and fingers, ensuring precise execution of the intended movement. Different speeds of movement correspond to variations in these neural signal patterns.
In the case of a patient with a forearm amputation, the brain’s ability to generate movement commands does not disappear after the amputation. The neural pathways remain intact, meaning that even though the hand is no longer physically present, the brain continues to issue movement commands for the missing fingers. These signals can be detected in the residual limb’s nerves and used to control a prosthesis in a manner similar to how the natural hand functioned before the amputation.
Integrating AI for Enhanced Functionality
The uniqueness of the NerveRepack prosthesis lies in its AI-driven approach. The use of plug-in electrodes allows the prosthesis to interpret and execute the patient’s neural commands in real time. Currently, no prosthetic device on the market fully integrates this type of direct neural control, making NerveRepack a pioneering effort in the field.
Training the AI Module: Learning from the Patient
The NerveRepack project is centered on an innovative concept—the bidirectional neural interface for prosthetic limbs. To ensure seamless integration, an AI module is being developed to recognize the patient’s brain commands and translate them into prosthetic movements.
The Training Process
For the AI module to function effectively, a training phase is required after the implantation of the electrodes. During this phase, the patient undergoes specialized learning sessions to adapt to the new prosthesis.
Use of a Smart Glove: The patient wears a glove on their healthy hand, which is equipped with flexion sensors to record data on finger movement—speed, acceleration, duration, and direction.
Simultaneous Motion Execution: The patient attempts to move both hands identically. While the healthy hand executes movements, the sensors collect data, and the implanted electrodes in the residual limb capture motor neural signals corresponding to those movements.
AI Training: Repeating each movement multiple times at varying speeds enables the AI module to correlate the collected neural data with specific commands, refining its recognition capabilities.
After sufficient training sessions, the AI module becomes capable of identifying the patient’s intended movements, allowing the prosthesis to execute commands naturally. Because the AI module is trained using data from the patient’s intact hand, the prosthesis' movements closely mimic natural hand function.
The Future of AI-Integrated Neuroprosthetics
The development of AI-driven prosthetic limbs represents a paradigm shift in neuroprosthetics, offering amputees an unprecedented level of control and sensory feedback. By integrating machine learning and neural interfaces, projects like NerveRepack push the boundaries of medical technology, paving the way for a future where prosthetic devices become extensions of the human body.
As research progresses, further refinements in AI algorithms, electrode design, and wireless communication will enhance the responsiveness and adaptability of prosthetic limbs. In a few years, these advancements could lead to commercial availability, offering hope to millions of amputees worldwide.
The NerveRepack project is not just about building a better prosthesis—it is about restoring autonomy and quality of life to individuals with limb loss. By harnessing the power of artificial intelligence and direct neural interfacing, this initiative is setting new standards in the field of prosthetics. With continued research and development, the dream of a fully integrated, patient-controlled prosthesis is drawing closer to reality, aligning human intention with artificial limb functionality.
References
[1] NBC News, Limb loss a grim, growing global crisis, http://haitiamputees.nbcnews.com/
[2] Rajah, J.K.; Chernicoff, W.; Hutchison, C.J.; Gonçalves, P.; Kopainsky, B. Enabling Mobility: A Simulation Model of the Health Care System for Major Lower-Limb Amputees to Assess the Impact of Digital Prosthetics Services. Systems 2023, 11, 22. https://doi.org/10.3390/systems11010022
[3] Advanced Amputee Solutions LLC, Amputee Statistics You Ought to Know, 2022, http://www.advancedamputees.com/amputee-statistics-you-ought-know
