Oct 06 2022

3D Printing Implantable Computer Chips

This is definitely a “you got chocolate in my peanut butter” type of advance, because it combines two emerging technologies to create a potential significant advance. I have been writing about brain-machine interface (or brain-computer interface, BCI) for years. My take is that the important proof of concepts have already been established, and now all we need is steady incremental advances in the technology. Well – here is one of those advances.

Carnegie Mellon University researchers have developed a computer chip for BCI, called a microelectrode array (MEA), using advanced 3D printing technology. The MEA looks like a regular computer chip, except that it has thin pins that are electrodes which can read electrical signals from brain tissue. MEAs are inserted into the brain with the pins stuck into brain tissue. They are thin enough to cause minimal damage. The MEA can then read the brain activity where it is placed, either for diagnostic purposes or to allow for control of a computer that is connected to the chip (yes, you need wired coming out of the skull). You can also stimulate the brain through the electrodes. MEAs are mostly used for research in animals and humans. They can generally be left in the brain for about one year.

One MEA in common use is called the Utah array, because it was developed at the University of Utah, which was patented in 1993. So these have been in use for decades. How much of an advance is the new MEA design? There are several advantage, which mostly stem from the fact that these MEAs can be printed using an advanced 3D printing technology called Aerosol Jet 3D Printing. This allows for the printing at the nano-scale using a variety of materials, included those needed to make MEAs. Using this technology provides three advantages.

First, the new MEAs can be rapidly customized. This means that if a researcher needs a specific MEA design for a specific experiment, with this technology they can have it in a matter of days. You can also customize for a specific patient, matching the MEA to their anatomy. Second, these MEAs can be three-dimensional. Current chips, like the Utah array, are two-dimensional. They are basically flat, like a regulat computer chip. But the 3D printing approach allows for the electrode pins to be of various lengths, the longer pins being inserted to a greater depth in the brain tissue, allowing for a functionally 3D array. Third, the nano-scale printing allows for a greater density of electrodes, by about an order of magnitude greater than existing MEAs.

This is a great example of something I discuss in my new book, which was released last week – The Skeptics’ Guide to the Future (and which I am now shamelessly plugging). When we think about how any one technology will advance, we also have to put this into context of how other technologies will be advancing alongside it. It is a common mistake of past futurists that they think of each technology largely in isolation. You always have to think – by the time we have this advance I am talking about, what other advances will have occurred? Will these other potential advances render the technology obsolete, change the nature of the problem we are trying to solve, or perhaps have a synergistic effect? The new 3D printed MEA is an example of the potential synergy between technologies – advances in additive manufacturing technology leads to advances in brain-machine interface technology.

Because this basic MEA design, basically electrodes on a chip, can only be left in place for about a year, they are mostly used for research or for diagnostic purposes. They are not a permanent therapeutic option, however. The main problem is that the brain pulsates slightly with each heart beat, and this causes the tiny electrodes to move a bit with respect to brain tissue. This causes two main problems. It can shift the relation between the MEA and the brain tissue. It also causes the formation of scar tissue, which eventually blocks the electrical connection between the brain and the MEA.

What we need for long term BCI is flexible MEAs. This will allow the electrode array to pulsate with the brain, keeping it from moving and reducing scar tissue. This now combines with a third type of emerging technology – flex-tech. Flexible electronics are already a thing, and we can print our 3D MEAs potentially out of flexible material. This is almost certainly where the BCI technology is eventually heading, especially if we want to use it for things like controlling robotic limbs.

The other technological advances we need for optimal BCI is to have some way for the electrodes to power themselves through ambient energy. They could, for example, tap into that pulsating brain to generate a small amount of electricity. Finally, we need to connect these MEAs to the outside of the skull wirelessly. Having wires come out of your head is not ideal, and not a permanent solution. This raises concerns of being able to hack into the MEA, so security would have to be absolutely tight. Flexible, self-powered, wireless MEAs could allow for long term control of electronic devices and robotic limbs. Clever scientists will likely think of other advances as well, but these are essential.

The fun part is imagining where this technology will lead in the distant future. In a sequel to 2001  – 3001: The Final Odyssey by Arthur C. Clarke, it is standard procedure for every human to have a “brain cap” installed. Essentially the top part of your skull is removed and replaced by a supercomputer which then interfaces to your brain through millions of micro-wires. The brain cap becomes a powerful third hemisphere of your brain, giving each person fantastically advanced artificial intelligence capability. We can also imagine “Matrix” scenarios where a dense-enough BCI allows for existence in a virtual world, a form of Neural-Reality. This technology also allows for humans to become cyborgs, with only their brains as an organic component.

But long before these scenarios play out, we will at least be able to replace lost limbs with reasonable robotic prosthetics, or restore some function to paralyzed limbs.


No responses yet