Thanks to the shrinking size of electronics, researchers have been exploring increasingly sophisticated implantable devices, paving the way for new prosthetics and brain-machine interfaces. But a big challenge has been how to deliver power to electronic components embedded within the body.
Now electrical engineers at the University of Washington have developed an implantable neural sensing chip that needs less power. Other wireless medical devices, such as cochlea or retinal implants, rely on inductive coupling, which means the power source needs to be centimeters away. The new sensor platform, called NeuralWISP, draws power from a radio source up to a meter away.
The device contains a microprocessor powered by a commercial radio-frequency reader that doubles as a data-collection device. The same equipment is used to power and read information from radio-frequency identification (RFID) tags. In experiments, the researchers used the new device to sense central nervous system activity in a moth in order to study its locomotion.
There have been some advances in reducing the size of neural implants recently, but the majority of implantable devices are still relatively cumbersome. These devices typically require multiple components–such as a clock for timing operations and an antenna for communication and power-harvesting–that are quite large compared to the transistors on the microcontroller, says Brian Otis, professor of electrical engineering at the University of Washington and lead researcher on NeuralWISP.
“You can have millions of transistors on a chip less that’s less than a cubic millimeter in volume, but the problem is with the extra parts,” says Otis. “Our goal is to shrink everything onto a single chip and reduce the power consumption of these components so that the chip can be wirelessly powered.”
The NeuralWISP is a collection of smaller, more low-power components, such as a specialized signal amplifier, on a circuit board just over two centimeters long. A future version will integrate all components onto a single chip that’s one millimeter by two millimeters in size. The circuitry converts usable power from the reader–roughly 430 microwatts–to a voltage that can turn on the microcontroller. This microcontroller, in turn, controls the sensor and its timer, and runs instructions that allow data to be sent back to the reader.
One of the main ways to save power, says Otis, was to reduce how often the sensor measured electrical signals produced by neurons. The researchers programmed the microcontroller to “wake up” when an electrical spike occurred, and record only the signals that were above a certain threshold. “Neuroscientists are interested in the spike rate,” says Otis. “We don’t digitize the entire brain waveform.”
In addition to a handful of low-power circuit design considerations, researchers built a small signal amplifier that boosts the electrical signal from neurons while minimizing electrical noise. For this, they split the incoming signal into two parts. The amount of incoming electricity from neural activity is the same, but by splitting it between a pair of transistors within the circuit, the amount of noise is cut in half.
In the moth experiment, researchers tested the battery-free system by collecting data on electrical signals from the moth’s wing muscles. The tests showed the frequency with which the moth flapped its wings. The results are published in the journal IEEE Transactions on Biomedical Circuits and Systems; the researchers also discussed the work at a summit on Wireless Identification and Sensing Platforms (WISP) in Berkeley, CA, on Tuesday. The current system is too large to allow the moth to fly freely, but an upcoming chip, which will be presented in February, is small enough to enable unencumbered flight, says Otis.
“Most implantable devices have used lower frequencies,” says Josh Smith, principal engineer at Intel, and organizer of the WISP Summit. A lower frequency also means that the devices must be read at close range, he adds. Using commercial RFID readers, says Smith, allows the device to be powered and data to be read from further away. However, he says it’s still an open question whether the antenna will maintain the long range once implanted in animal tissue, because the signal might be absorbed. “Measuring moths is a good fit for this approach, since the antenna does not have to go inside the animal’s tissue,” he says.
Arto Nurmikko, professor of engineering at Brown, agrees, saying that it’s useful to measure neural activity in moths, “but the real challenges and application potential emerge in work with primates.”
The dark secret behind those cute AI-generated animal images
Google Brain has revealed its own image-making AI, called Imagen. But don't expect to see anything that isn't wholesome.
The hype around DeepMind’s new AI model misses what’s actually cool about it
Some worry that the chatter about these tools is doing the whole field a disservice.
The walls are closing in on Clearview AI
The controversial face recognition company was just fined $10 million for scraping UK faces from the web. That might not be the end of it.
This horse-riding astronaut is a milestone in AI’s journey to make sense of the world
OpenAI’s latest picture-making AI is amazing—but raises questions about what we mean by intelligence.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.