A research team from Fudan University in China has successfully developed artificial neurons that closely mimic the behaviour of biological brain cells. The breakthrough comes amid growing global demand for computing systems that deliver higher performance while consuming less energy, driven by the rapid advancement of artificial intelligence (AI) and machine learning technologies.
Machine learning algorithms are inspired by biological neural networks, and researchers have long sought to replicate the brain’s structure and functionality. The newly developed “artificial neurons” achieve this by incorporating dynamic connections whose strength changes over time, simulating the brain’s ability to learn and adapt.
The innovation relies on two key components: specialised “cells” that store electrical charge to imitate the membrane potential of biological neurons, and a “switching mechanism” that generates signals resembling neural spikes. This design allows the system to process information in a way that parallels natural brain activity.
To test the concept, researchers constructed a compact 3×3 network of artificial neurons and evaluated its response to varying light intensities—an experiment designed to simulate human vision under different lighting conditions. Results demonstrated that the artificial neurons performed strongly in tasks such as computer vision and image recognition, while maintaining remarkable energy efficiency.
The development marks a significant step toward building next-generation neuromorphic systems—computing architectures inspired by the human brain. By combining adaptability with low energy consumption, these artificial neurons hold promise for future applications in machine vision, pattern recognition, and advanced AI systems.
