Losing a hand is a highly disabling event – one that affects nearly 3 million people worldwide. To help mitigate the impact that losing a hand has on one’s quality of life, amputees can use a prosthetic arm. However, these are only as useful as the level of real-life sensation and dexterity that they can provide – and most of the models currently on the market fail to offer either. Consequently, many amputees reject them due to their lack of sensory feedback and the cognitive workload required to use them.
This could soon change, thanks to several new tools for assessing the neural and muscle activity of state-of-the-art bionic hand prosthesis being advanced by the BIREHAB project.
“For a limb to feel natural, we need to be able to control it, but also receive and process all the information from the outside world,” says Fiorenzo Artoni, BIREHAB project coordinator and former Marie Skłodowska-Curie Fellow at the École Polytechnique Fédérale de Lausanne.
Focusing on natural control and sensory feedback integration, Artoni developed tools that can assess the ‘naturalness’ of a prosthetic hand’s tactile sensory feedback. This information can then be used to improve the control of hand prostheses, ultimately allowing for the patient to ‘feel’ as with a real hand.
The first goal of the project was to lay the groundwork for building a robust, myoelectric-controlled prosthesis that can determine how tactile stimuli are perceived by the patient. Artoni thus designed a series of electroencephalography (EEG)-based tests to evaluate the brain’s electrical activity.
Next, Artoni studied the prosthesis’ ability to naturally react to stimuli. To do this, he conducted an experiment where amputees and non-amputees received small electric pulses to their forearms. This allowed him to see whether the pulse elicited a tingling sensation in the prosthetic hand and, by watching the subject’s brain activity, see whether the reaction was similar to what was produced with a real hand in healthy subjects.
“The results showed clear differences between stimulations on the forearms that are felt only on the forearm and stimulations on the forearms that are felt on a prosthetic hand,” explains Artoni. “Interestingly, the neural correlates were strikingly similar to those achieved with real tactile stimulation to non-amputees, which we delivered by sliding gratings beneath their fingers.”
A better way of measuring muscle activity
Of course, there is another side of the story. A myoelectric-controlled prosthesis needs to be controlled using electrical signals generated by the amputee’s own muscles in as natural a way as possible. This, however, requires recording and analysing, in real time, the residual muscle activity generated at the forearm and sending control signals to the prosthesis according to the patient’s intention.
The main issues here are the long setup times and the difficulty of locating forearm muscles with enough precision to place the sensors correctly. “A possible solution is to cover the whole forearm with sensors,” he says. “But what is the optimal number of sensors needed to achieve good decoding performance while limiting hardware complexity?”
To find out, Artoni studied how the different number of channels influenced the quality of hand gesture decoding. “Based on this, I developed an electromyogram sleeve that drastically reduces recording and setup times, as well as the need to locate each muscle individually,” he adds.
An incredible opportunity for growth
Based on the success of these experiments, Artoni is currently working to patent BIREHAB’s hardware and software solutions. He has also co-authored 15 papers published in various professional journals and has presented at a range of conferences and workshops.
“The Marie Skłodowska-Curie fellowship really represents an incredible opportunity for growth and professional development,” concludes Artoni. “I’ve received very positive feedback on my work and found this research experience incredibly enjoyable and fulfilling.”