A human brain cell was put on a chip and can now do math

One of the organoids in the experiment conducted by linking brain cells to computer chips. Credit: Cai et al., Nat. Electron., 2023
One of the organoids in the experiment. Credit: Cai et al., Nat. Electron., 2023
Share this:

Our brain is powerful, more powerful than any computer. Its 86 billion neurons help it process information and our environment by emulating memory chips and processors, all aided by over a quadrillion synapses. Due to this immense power and capacity, efforts are continuously being made to make computers perform more like our brains. And now a new innovation has taken a step forward, by integrating human brain tissue with electronics.

This new attempt is called Brainoware and it involves using human brain tissue grown in a lab. It was undertaken by a team led by engineer Feng Guo of Indiana University Bloomington, who compelled human pluripotent stem cells into developing into different types of brain cells that eventually formed into mini-brains called organoids that also had connections and structures. Rest assured that these are not true brains but rather just tissues. They do not function like our brains and have no thoughts, feelings or emotions. Guo and team followed the ethics guidelines in the development of Brainoware.

“Here we report an artificial intelligence hardware approach that uses adaptive reservoir computation of biological neural networks in a brain organoid. In this approach—which is termed Brainoware—computation is performed by sending and receiving information from the brain organoid using a high-density multielectrode array. By applying spatiotemporal electrical stimulation, nonlinear dynamics and fading memory properties are achieved, as well as unsupervised learning from training data by reshaping the organoid functional connectivity.”, says the study team.


Left to right, top: Human brain organoids at 7 days, 14 days, 28 days, and several months; Bottom, left to right: 1 month, 2 months, 3 months. Credit: Cai et al., Nat. Electron., 2023

Left to right, top: Human brain organoids at 7 days, 14 days, 28 days, and several months; Bottom, left to right: 1 month, 2 months, 3 months. Credit: Cai et al., Nat. Electron., 2023


So, basically, Brainoware comprises of the aforementioned brain organoids connected to high-density microelectrodes, to which information is transported via electrical simulation. Standard computer components were used to handle both data input and output. These components were specially adapted to work with the organoid, allowing the output layer to interpret neural signals and draw conclusions or make predictions.

Information is processed and answers are revealed in the form of neural activity. Guo and team then gave it various tasks such as speech recognition and math problems. To showcase Brainoware’s potential, researchers fed it 240 audio clips of eight men saying Japanese vowels, then challenged it to pick out one specific voice. With just two days of training, this “brain-on-a-chip” achieved a remarkable 78% accuracy in speaker identification.

Further testing its limits, the researchers tasked Brainoware with predicting a chaotic Hénon map. After four days of unsupervised learning, Brainoware outperformed a basic neural network, showcasing its potential for handling complex dynamics.

“We illustrate the practical potential of this technique by using it for speech recognition and nonlinear equation prediction in a reservoir computing framework”, highlight the team.

Although Brainoware was a little less accurate than artificial neural networks with a long short-term memory unit, it achieved almost the same results in less than 10 percent of the training time.

“Due to the high plasticity and adaptability of organoids, Brainoware has the flexibility to change and reorganize in response to electrical stimulation, highlighting its ability for adaptive reservoir computing,” the researchers write.

While there are still significant limitations to this innovation, such as keeping the organoids alive and functioning, as well as the required power consumption levels, it still has foudnational implications for not only for the future of computing but also increase understanding of how our brain works.

The research has been published in Nature Electronics.


Become a Patron!


Buy me a coffee


I am a Chartered Environmentalist from the Royal Society for the Environment, UK and co-owner of DoLocal Digital Marketing Agency Ltd, with a Master of Environmental Management from Yale University, an MBA in Finance, and a Bachelor of Science in Physics and Mathematics. I am passionate about science, history and environment and love to create content on these topics.

Free Email Updates
We respect your privacy.