分类： 计算机科学 >> 计算机软件 提交时间： 2025-03-01
摘要： In cognitive and computational neuroscience, the challenge of generalizing EEG-based models across subjects stems from inter-individual variability and low signal-to-noise ratios (SNR). Traditional EEG analysis methods often fail to account for these differences, resulting in models that perform well on training data but poorly on new subjects. To address this, we propose SwinEEG, a novel individual-to-individual EEG conversion framework inspired by the Swin Transformer, a vision transformer known for its hierarchical architecture and self-attention mechanisms. SwinEEG leverages contrastive learning and image reconstruction to map neural representations between subjects by incorporating both image stimuli and source-subject EEG data. Experiments on the THINGS EEG2 dataset demonstrate improved inter-subject neural mapping performance, with an average conversion accuracy increase from 71.2 to 74.3 compared to prior methods. This approach highlights the potential of combining vision-inspired transformers and contrastive learning to mitigate individual differences and noise in EEG signals.
分类： 生物学 >> 植物学 >> 植物生理学 提交时间： 2016-05-03
Haibing Yang
Hui Wei
Guojie Ma
Mauricio S. Antunes
Stefan Vogt
Joseph Cox
Xiao Zhang
Xiping Liu
Lintao Bu
S. Charlotte Gleber
Nicholas C. Carpita
Lee Makowski
Michael E. Himmel
Melvin P. Tucker
Maureen C. McCann
Angus S. Murphy
Wendy A. Peer
摘要： Conversion of nongrain biomass into liquid fuel is a sustainable approach to energy demands as global population increases. Previously, we showed that iron can act as a catalyst to enhance the degradation of lignocellulosic biomass for biofuel production. However, direct addition of iron catalysts to biomass pretreatment is diffusion-limited, would increase the cost and complexity of biorefinery unit operations and may have deleterious environmental impacts. Here, we show a new strategy for in planta accumulation of iron throughout the volume of the cell wall where iron acts as a catalyst in the deconstruction of lignocellulosic biomass. We engineered CBM-IBP fusion polypeptides composed of a carbohydrate-binding module family 11 (CBM11) and an iron-binding peptide (IBP) for secretion into Arabidopsis and rice cell walls. CBM-IBP transformed Arabidopsis and rice plants show significant increases in iron accumulation and biomass conversion compared to respective controls. Further, CBM-IBP rice shows a 35% increase in seed iron concentration and a 40% increase in seed yield in greenhouse experiments. CBM-IBP rice potentially could be used to address iron deficiency, the most common and widespread nutritional disorder according to the World Health Organization.
点击量
通过
点击量
下载量
评论
待评议
