A Biomedical Microelectronics Challenge for Electronics Designers

By: Rudy Lauwereins, Vice President, Smart Systems Technology Office, IMEC

My father-in-law has Parkinson’s disease. Apart from the typical tremors, he also suffers from muscle rigidity. A while ago, deep brain stimulation (DBS) electrodes were implanted in his brain to stimulate the affected zones. These electrodes allow him to move normally again, but they come at the cost of some unpleasant side effects. That is mainly because the technology in use today is not able to stimulate precisely enough. It is as if you were firing a flash of lightning in a room to light a candle. Microelectronics and chip technology could help to improve DBS, but we still have a lot to learn. For example, how to design electronics that work without fail in a liquid, nonhomogeneous environment such as our brains.

 

DBS works by stimulating regions of the brainstem with electrodes. During the implantation, two holes are drilled in the skull trough, through which a wire is pushed, each wire with four millimeter-sized electrodes at its tip. The doctors push the wires through the brain tissue until the electrodes hit the right zone. Once these electrodes are implanted, they can send electrical impulses to the affected zone, just like a pacemaker does for the heart. The impulses are generated by a battery powered pulse generator that is typically implanted below the collarbone. This stimulator generates pulses of 5 V with a frequency of approximately 100 Hz. The electronics can be adjusted, to calibrate the stimulation, optimise the suppression of symptoms and control side effects

 

Before the implantation, my father-in-law could no longer eat with a spoon; his rigid muscles and tremor made this impossible. After the operation, thanks to the stimulation, he was happily ladling his soup. With other patients, you see that they “freeze up” every time they try to walk through a doorway. They cannot summon up the muscle coordination that is needed and they come to a total standstill in front of the doorway. If, in such an instance, you switch on their DBS, they suddenly “thaw” and walk through the doorway as if nothing happened.

 

But DBS may also cause unwanted side effects. In the case of father-in-law these were hallucinations and a loss of inhibitions. The causes of this are, first, that the stimulation is macroscopic; as well as the brain zone that is targeted, the stimulation also hits other zones. Second, it is not possible to direct the electrodes precisely enough through the brain tissue. Sometimes they are located a full centimetre away from the targeted neurons. Third, current DBS is an open-loop system. You cannot measure the result of the stimulation, thus you cannot adapt it to what is needed at any moment.

 

 

For each of these problems, microelectronics has a solution to offer. We can make much smaller electrodes, eventually as small as the neurons themselves. With these, we could stimulate a much smaller group of brain cells, without side effects on neighbouring cells.

 

But next comes the problem of locating the electrodes at exactly the right place. If you cannot do that, you have to stimulate a larger zone anyway. The solution is beam forming. By using an array of electrodes and modifying the phase differences between the pulses of the electrodes, the pulses can be intensified in one direction and attenuated in another direction. With this type of DBS, the stimulation can be directed.

 

A third challenge is to make a closed-loop DBS. Today’s DBS stimulates a neuron with a pulse of 5 V, normal electrical resting potential of that neuron is only approximately 60mV. If you wanted to measure the result of that pulse on the neuron, your measuring equipment would be totally blinded by the 5 V. There is a promising solution for this problem.

 

 

At the Design, Automation and Test in Europe (DATE) 2009 conference, IMEC presented a prototype DBS that can stimulate and measure at the same time, which has electrodes of 10 µm.

 

IMEC engineers design electronic applications with the help of simulation. For a wireless sender–receiver for example, we make an electronic model of the sender, use that to send signals through a model of the atmosphere and then see how we can recuperate the signal on a model of the receiver. The tool we use for doing that is MATLAB.

 

But a traditional design method like this does not work in the fluidic, heterogeneous environment of the brain. The IMEC DBS prototype has used a tool for element modelling and we coupled that to our hardware modelling flow. That allowed us to simulate precisely how we have to steer the electrodes to arrive at a directional stimulation in an environment like that of the human brain.

 

Designing this application was, in the larger perspective, an exercise in designing heterogeneous electronics. In this case electronics combined with a biological medium. In the future, we are also going to make designs that combine CMOS with, for example, mechanics (MEMS), or with a biological surface such as a cell. We have made a dedicated design flow for one heterogeneous application, but our goal is eventually to arrive at a generic design flow for heterogeneous electronics.