Implant Manufacturing Faces Paradigm Shift

Prof. Lacour's career can be summarized as a quest to meet the formidable challenge of combining biology with electronics. In her case, that means coupling soft materials with rigid components and enabling ions in the human body to communicate with electrons in circuits. Since joining EPFL in 2011, Lacour has been conducting research at the crossroads of materials science and electrode miniaturization - a job that requires inventiveness and creativity. She is a co-founder and member of the Neuro X Institute, which brings together experts in neuroscience, neuroengineering and neuroinformatics to explore medical applications. She recently stepped down as the director of Neuro X when she was appointed to her new role as EPFL's Vice President for Support to Strategic Initiatives, which she took up on 1 January 2025.

Twenty years ago, would you have imagined that we'd be where we are today - paraplegic patients are able to walk again with soft implants, and we can control computers with our thoughts?

Probably. Neurotechnology is a field where the futuristic systems portrayed in science fiction are now becoming reality. There's been a lot of buzz around neurotechnology recently, but we shouldn't forget that one of the first neurotechnological systems - the cochlear implant - was introduced over 40 years ago and has helped over a million patients worldwide. So we've been able to modulate the nervous system for a while. What's new today, however, is the development of brain-computer interfaces (BCIs): closed-looped systems that decode brain signals reflecting a user's intention and translate them into actions - such as activating a muscle or controlling a prosthesis - thereby establishing a direct link between the brain and an external device.

So there's been considerable progress.

Yes, but also a lot of hype. Few of the implantable BCIs that exist today are ready for use as medical devices. Most of them are in the testing or clinical-trial stage. But the pace of neurotechnology development has accelerated thanks to advancements in fabrication and miniaturization methods, which have improved the precision and capabilities of BCIs.

Has the development of soft electrodes also played a role?

Soft materials have advantages in terms of biointegration and disadvantages in terms of manufacturing and stability. Our aim is to develop implants that will result in minimal inflammation in patients, based on the idea that an implant with physical properties similar to those of human tissue should theoretically trigger fewer adverse reactions. But these efforts are still in the preclinical trial stage or are just starting to be tested on humans. We don't have enough data on how stable they will be over time, and therefore on their efficacy. It's a work in progress.

Is it important to fit as many electrodes as possible onto a single device?

It depends on what the goal is. We need to make a distinction between devices for therapeutic applications and those for obtaining a better understanding of the human brain. Our brain contains billions of neurons, and scientists who want to study how it all works need systems with a very high electrode density, so that they can generate precise maps of brain activity. On the other hand, such a high density isn't necessary for most therapeutic applications. To control a prothesis, for example, all you basically need is to capture a brain signal that provides broad information about a neural network. BCIs don't need a billion electrodes to control an exoskeleton with a great deal of precision or to stimulate the spinal cord. From that perspective, the information collected by noninvasive or minimally invasive systems such as EEGs or electrocorticography (ECoGs) is enough.

The biggest challenge in designing high-electrode-

density systems is to be able to miniaturize the power supply, extend the battery life and develop programs that can crunch through the reams of data they collect in real time. Engineers are working on integrating algorithmic processes right into chips so that systems can decode signals locally and transmit only the relevant data. But this approach is still in the exploratory stage.

When do you think these kinds of therapies will be more widely available?

It could be really soon, or it could take a while. Experience has shown that it generally takes 15 to 25 years before a device is accepted as a therapeutic option for humans - one that's prescribed in hospitals and covered by health insurance. How quickly neurotechnology is adopted will also depend on how easily manufacturers are able to set up mass production facilities that can both produce large volumes and lower the unit cost - a crucial hurdle in making such technology accessible.

Will noninvasive technology like EEGs be adopted more rapidly?

Here too, it depends on the research goal or therapeutic aim. EEGs are widely used in diagnostics and to study sleep disturbances. But they have a relatively low spatio-temporal resolution. For instance, owing to their low spatial resolution, doctors can't get a granular view of a patient's brain activity, such as when the patient tries to move individual fingers.

Besides paraplegia patients with motor problems, who else could benefit from neural implants?

Motor problems are the starting point for neural implants because they create a direct link between neurostimulation and a specific action, such as to controlling a prothesis or exoskeleton. Other medical applications include facilitated communication, adaptative neurostimulation to treat epileptic seizures and Parkinson's tremors, and neurofunctional rehabilitation for people who have suffered a stroke. A research team in the US recently enabled a stroke victim to speak again thanks to the use of a BCI.

Neurotechnology is a field where the futuristic systems portrayed in science fiction are now becoming reality. But we shouldn't forget that one of the first neurotechnological systems - the cochlear implant - was introduced over 40 years ago and has helped over a million patients worldwide.

What about applications beyond the brain?

You're right - the nervous system isn't composed only of the brain. The peripheral nerves are important too. Here, cochlear implants are an example of a successful technology. Scientists have also developed implants for the vagus nerve and other peripheral nerves to treat metabolic disorders, but these devices are still in the research stage.

Other parts of the body could also be regulated or monitored with implants. In this case, the implants wouldn't be electrodes but rather chemical or biochemical sensors. Yet the basic components of the overall system - a battery that powers a connected electronic device coupled to a sensor network - would be the same.

Looking further out, we'll probably see a form of hybrid technology emerge that consists of a device interacting directly with the nervous system to restore or repair lost functions, or to gradually stimulate our natural biological mechanisms to repair them.

It sounds like neurotechnology is an extremely promising development for treating many diseases - somewhat like genetics was in the 2000s. Is that right?

It's true that the excitement surrounding neurotechnology today is similar to what we saw with genetics in the 2000s. Neurotechnology can be used to restore lost functions by interacting directly with the brain. The devices are becoming increasingly miniaturized, precise and customizable, opening up new horizons for personalized medicine. That said, there are major technical, ethical and regulatory obstacles to overcome. Scientists have made great strides, but the translation into clinical applications has been slow and modest.

What could we do to speed things up?

Neurotechnology is currently in a phase of rapid development. Numerous research groups are studying new materials, designs and production methods for making miniaturized BCIs that are more precise, stable and reliable. By integrating microelectronics and soon AI into these systems, we should be able to take a major step forward in enabling the efficient, real-time decoding and encoding of neural data.

A key factor is the need for cross-disciplinary collaboration to drive advancements in neurotechnology along with structured initiatives involving both the public and private sectors, which will help scientists work through the critical steps of development, testing and eventual use with patients.

In my role as EPFL's Vice President for Support to Strategic Initiatives, I'm advocating for this kind of holistic approach where EPFL brings together scientists and engineers from several disciplines to address major societal challenges.

How is EPFL positioned in neurotechnology?

Neurotechnology is a field where it's essential to combine methods from life science, engineering, AI and medicine. Around ten years ago, EPFL opened the Center for Neuroprosthetics with five founding members with the goal of pooling local skills to create an ecosystem. Then in 2022, three EPFL schools - Life Sciences, Engineering and Computer Science, teams up to open the Neuro X Institute mostly based at Campus Biotech in Geneva. Today Neuro X is home to the complementary and crucial expertise that was based at the Center for Neuroprosthetics. The Campus Biotech ecosystem and dedicated testing facilities encourage a cross-disciplinary approach and facilitate translational research. EPFL also has R&D partnerships with Lausanne University Hospital, Geneva University Hospitals and Valais Hospital, and receives support from several foundations, including the Bertarelli, Wyss and Defitech Foundations, which assist us in particular with translational research.

Things are harder for neurotechnology startups, many of which have failed.

This kind of technology requires such a huge investment, and over such a long time period, that you've got to have really solid footing to start a business and keep it going. Many startups fail because the approval process is extremely lengthy, and they often don't have a manufacturer to produce their device, or therapeutic examples to demonstrate, or acceptance from health insurers.

But a paradigm shift is under way. We're transitioning from a very traditional market based on proven, long-standing technology to one that's open to new designs and therapeutic approaches. My hope is that an agile, ambitious environment will emerge that promotes the adoption of neurotechnology. Switzerland and Europe in general are very well positioned in neurotechnology R&D, but we'll need concrete investments similar to what's being done in the US if we want to take lasting steps forward.

Could this kind of technology also augment individuals' capabilities?

Maybe. I'm focused on advancements that can bring tangible benefits to patients today, rather than more speculative considerations.