The field of neural decoding and brain-computer interfaces (BCIs) is rapidly advancing, with a focus on developing more accurate and efficient methods for decoding neural activity and controlling devices. Recent research has explored the use of deep learning models, such as Cortical-SSM and NeurIPT, to improve the accuracy of neural decoding and BCIs. These models have shown promising results in decoding electroencephalogram (EEG) and electrocorticogram (ECoG) signals, and have the potential to be used in a variety of applications, including communication assistance and rehabilitation support for patients with motor impairments. Additionally, researchers have been investigating the use of transfer learning and domain adaptation to improve the performance of BCIs across different subjects and sessions. Noteworthy papers in this area include Cortical-SSM, which proposes a novel architecture for EEG and ECoG motor imagery decoding, and NeurIPT, which develops a foundation model for neural interfaces with a pre-trained transformer. Overall, the field of neural decoding and BCIs is rapidly advancing, with a focus on developing more accurate and efficient methods for decoding neural activity and controlling devices.
Advances in Neural Decoding and Brain-Computer Interfaces
Sources
Cortical-SSM: A Deep State Space Model for EEG and ECoG Motor Imagery Decoding
Transfer Orthology Networks
A Real-Time BCI for Stroke Hand Rehabilitation Using Latent EEG Features from Healthy Subjects
WaveNet's Precision in EEG Classification
Seeing Through the Brain: New Insights from Decoding Visual Stimuli with fMRI
NeurIPT: Foundation Model for Neural Interfaces
Foundation and Large-Scale AI Models in Neuroscience: A Comprehensive Review
Finding Manifolds With Bilinear Autoencoders
Fly-CL: A Fly-Inspired Framework for Enhancing Efficient Decorrelation and Reduced Training Time in Pre-trained Model-based Continual Representation Learning
Uncovering Brain-Like Hierarchical Patterns in Vision-Language Models through fMRI-Based Neural Encoding
TopSeg: A Multi-Scale Topological Framework for Data-Efficient Heart Sound Segmentation
Muscle Anatomy-aware Geometric Deep Learning for sEMG-based Gesture Decoding
Atlas-based Manifold Representations for Interpretable Riemannian Machine Learning
BrainMCLIP: Brain Image Decoding with Multi-Layer feature Fusion of CLIP
Coupled Transformer Autoencoder for Disentangling Multi-Region Neural Latent Dynamics
HybridSOMSpikeNet: A Deep Model with Differentiable Soft Self-Organizing Maps and Spiking Dynamics for Waste Classification
A Scalable, Causal, and Energy Efficient Framework for Neural Decoding with Spiking Neural Networks
MEIcoder: Decoding Visual Stimuli from Neural Activity by Leveraging Most Exciting Inputs