Thursday, 14 December 2023 – 10:04 WIB
Jakarta – In an effort to increase the computing power of artificial intelligence (AI), researchers have combined existing machine learning with sophisticated 3D models of human mini-brains made from various types of laboratory-grown brain tissue.
Reported LIVE Techno from Live ScienceThursday, December 14, 2023, These miniature models of the brain, known as brain organoids or “minibrains”, have existed in various forms since 2013. However, they have never been exploited as a way to improve AI.
Artificial intelligence (AI).
The new research uses more traditional computing hardware to feed electrical data into the organoid and then decomposes the activity of the organoid to produce output, so that the organoid serves only as a “middle layer” of the computational process.
While this method is far from mimicking the actual structure of the brain or how it works, it could be a first step towards creating biocomputers, which would borrow tricks from biology to make them more powerful and energy efficient than traditional computers.
It may also provide more insight into how the human brain operates and how the human brain is affected by neurodegenerative conditions, such as Alzheimer’s and Parkinson’s disease.
For the new study, published in the journal Nature Electronics, the researchers used a technique called reservoir computing; in this context, organoids function as “reservoirs.” In such a system, the reservoir stores information and reacts to the information entered.
An algorithm learns to recognize changes triggered in the reservoir by different inputs and then translates these changes as outputs.
Using this framework, the researchers plugged brain organoids into this system by supplying them with electrical input delivered via electrodes.
“Essentially, we can encode information – such as images or audio information – into temporal-spatial patterns of electrical stimulation,” said study co-author Feng Guo, a professor of intelligent systems engineering at Indiana University Bloomington.
Artificial intelligence technology mimics the way the human brain works.
In other words, organoids respond differently depending on the timing and spatial distribution of electricity from the electrodes. The algorithm learns to interpret the organoid’s electrical response to those stimuli.
Although brain organoids are much simpler than actual brains – they are essentially a small ball of brain cells – they have the ability to adapt and change in response to stimuli.
The responses of different types of brain cells, cells at different stages of development, and brain-like structures in organoids provide a rough analogue to the way our brains change in response to electrical signals. These changes in the brain drive our ability to learn.
Using this unusual hardware, the researchers trained their hybrid algorithm to solve two types of tasks: one related to speech recognition and one related to mathematics.
In the first task, the computer demonstrated about 78 percent accuracy in recognizing Japanese vowel sounds from hundreds of audio samples. And it is quite accurate in solving math tasks, but slightly inferior to traditional types of machine learning.
First Time Brain Organoids Used
Artificial intelligence illustration.
This research marks the first time brain organoids have been used with AI, but previous studies have used this type of lab-grown neural tissue in a similar way.
For example, scientists have wired brain networks with a form of reinforcement learning, a type of machine learning that may have more in common with the way humans and other animals learn than reservoir computing.
Future research could try to combine brain organoids with reinforcement learning, said Lena Smirnova, an assistant professor of environmental health and engineering at Johns Hopkins University who co-authored a commentary on the new study.
One of the advantages of creating biocomputers is energy efficiency, as our brains use much less energy than today’s advanced computing systems. However, Smirnova said that it may be decades before technology like this can be used to create biocomputers that can be used in general.
Although organoids don’t yet come close to replicating a complete human brain, Smirnova hopes the technology will give scientists a better understanding of how the brain works, including in diseases like Alzheimer’s.
Replicating brain structure (with organoids) and function (with computing) could allow researchers to better understand how brain structure is related to learning and cognition, for example.
Like organoids in general, this computational system is expected to help replace drug testing on animals, Smirnova added, which raises ethical issues and doesn’t always produce useful results because animals are so different from humans.
Incorporating organoids derived from human brain tissue into drug testing could help close that gap.
2023-12-14 03:04:02
#Scientists #Create #Mini #Brains #Computers
