How does it work

The entire system consists of brain organoids (clusters of human cells that mimic organs) and conventional electronics. To do this, the researchers placed “an organoid on a plate with thousands of electrodes to connect the brain to electrical circuits.” Circuits that “talk” to the brain organoid “translate the information they want to input into the form of electrical impulses.” Scientists from Indiana University converted the information into electrical impulses and transferred them to the organoid. Sensors recorded the brain tissue’s response, and a machine learning algorithm deciphered it.

It was used for voice recognition to test the system’s capabilities. 240 audio files containing the voices of eight different people were converted into electrical signals and sent to the organoid for analysis. The mini-brain responded differently to each sound and showed its own pattern of neural activity in each situation. Artificial intelligence trained to interpret these reactions identified the person in the recording. After training, the system was able to recognize voices with 78% accuracy.

The researchers admit that there is still a long way to go before a full-fledged biological computer emerges, but their work has already confirmed some basic theoretical propositions. In previous experiments, only two-dimensional nerve cell cultures had been subjected to similar tests. This time, for the first time, three-dimensional brain organoids have overcome this task.

In practical terms Biocomputers could provide computers of the future with the speed and energy efficiency of the human brain. Because organoids can reproduce the architecture and function of a working brain in a way that simple cell cultures cannot, the technology can also be used to study brain processes and neurodegenerative diseases.