In a groundbreaking development, researchers from HSE University and the Artificial Intelligence Research Institute (AIRI) have harnessed the power of artificial intelligence to revolutionize neurofeedback. The team, led by Alexei Ossadtchi, achieved a remarkable 50-fold reduction in latency between changes in brain activity and the corresponding neurofeedback signal.
This achievement holds significant promise for advancing treatments related to attention deficit disorder and epilepsy. Published in the Journal of Neural Engineering, the study challenges existing limitations in neurofeedback and introduces a potential game-changer for patients undergoing neurological therapies.
...
... the research team turned to artificial intelligence, training neural networks with extensive datasets of individuals’ brain activity. The objective was to create a model capable of accelerating the detection of target signals amid the broader spectrum of brain activity. Various neural network architectures were tested, with the Temporal Convolutional Network (TCN) emerging as the most effective.
Implementing TCN allowed the construction of a filter that isolated rhythmic brain activity, leading to an unprecedented reduction in latency. The delay in presenting feedback signals, reflecting instantaneous alpha rhythm intensity, was slashed to just 10 ms. This represents a remarkable advancement, lowering the delay by approximately fifty-fold compared to conventional neurofeedback systems.
...
... The study’s findings ... provide a potential breakthrough in enhancing the efficacy of neurofeedback therapies. ...
