Japanese scientists have made a breakthrough by creating the world’s first images based on human brain activity using artificial intelligence (AI).
This new technology, called “brain decoding,” allows the visualization of what people perceive, such as objects and landscapes, from their brain signals.
A research team from the National Institutes for Quantum Science and Technology (QST) and other organizations worked on this project. They successfully generated rough images of a leopard, showing its mouth, ears, and spotted pattern. They also created images of other objects like an airplane with red lights on its wings.
The technology works by analyzing brain activity using functional magnetic resonance imaging (fMRI). During the study, participants viewed 1,200 different images of objects and landscapes.
Scientists measured their brain signals while they looked at these images. Then, they input these same images into AI to help it learn the relationship between the brain activity and the visual content.
This research could have practical applications in the medical and welfare fields. For example, it could help develop communication devices for people who cannot speak or assist in understanding how hallucinations and dreams work in the brain.
One of the researchers, Kei Majima, pointed out that while humans have long used devices like microscopes to see things invisible to the naked eye, this is the first time we have been able to look inside a person’s mind. The technology could allow us to better understand how the brain processes the world around us.
The team’s findings were recently published in the international scientific journal Neural Networks, marking a significant step forward in neuroscience and AI research.