Researchers showed neurons growing on a micron-scale CMOS bed of nails. "This is a major step forward because each electrode records only one neuron," said Wolfgang Eberle, a senior scientist here.

The chip lets researchers monitor the interactions of neurons over the course of as much as a month. Currently engineers use a microscope to manually insert a probe into a neuron, monitoring it for about a day before the cell dies.

Instrument companies are showing interest in the chip for automated testers they hope to design. Longer term, "pharma companies are looking over our shoulder," said Kris Verstreken, director of IMEC's bio-nano department.

Drug makers are interested in the technology as a way ease the job of testing new drugs, a process that can take as long as a decade and cost as much as $800 million. However, they want to see the device tested in tens of thousands of cultures over the course of months before they are willing to adopt it.

IMEC launched a neural engineering institute earlier this year to do just that. The partnership with a local university and bioengineering center aims to hire 50 researchers in a lab which will officially open next year.

"Once the chip is a proven as a research tool, then you can bring it to the Johnson & Johnsons of the world," said Verstreken. "Everyone wants to help pharma find new drugs, but no one else has these tools," he added.

The IMEC work is part of an industry-wide effort to move down from working with clusters or neurons to individual nerve cells to better understand the human brain and deal with neural diseases.

Verstreken gave a video tour of the levels of complexity in the human brain that researchers aim to surmount. He also sketched out his vision of moving to a 3-D device that grows nerve cells and their supporting cells in a more lifelike environment to achieve even more meaningful results about neural interactions.

"You need tools at the level of a micrometer due to the complexity of nerves and their dendrite nodes, each a computer in its own right," he said. "Ultimately, we hope to use these neural interfaces to reconnect parts of the brain damaged by tumors or trauma," he said.

It's early days for medical electronics efforts like this at IMEC which makes most of its revenues from partnerships on semiconductor process technology R&D. Medical electronics companies are not IMEC's core constituents and they are just coming around to a desire to do collaborative research.

"They don't even have a business model for this yet," said one IMEC executive.