Review Article Design and optimization of auditory prostheses using the finite element method: a narrative review Qianli Cheng, Han Yu, Junpei Liu, Qi Zheng, Yanru Bai, Guangjian Ni
Semantic understanding is central to advanced cognitive functions, and the mechanisms by which the brain processes language information are still being explored. Existing EEG datasets often lack natural reading data specific to Chinese, limiting research on Chinese semantic decoding and natural language processing. This study aims to construct a Chinese natural reading EEG dataset, TMNRED, for semantic target identification in natural reading environments. TMNRED was collected from 30 participants reading sentences sourced from public internet resources and media reports. Each participant underwent 400–450 trials in a single day, resulting in a dataset with over 10 hours of continuous EEG data and more than 4000 trials. This dataset provides valuable physiological data for studying Chinese semantics and developing more accurate Chinese natural language processing models.
26 February 2024 in IEEE Transactions on Cognitive and Developmental Systems
Cognitive Language Processing (CLP), situated at the intersection of Natural Language Processing (NLP) and cognitive science, plays a progressively pivotal role in the domains of artificial intelligence, cognitive intelligence, and brain science. Among the essential areas of investigation in CLP, Cognitive Signal Decoding (CSD) has made remarkable achievements, yet there still exist challenges related to insufficient global dynamic representation capability and deficiencies in multi-domain feature integration. In this paper, we introduce a novel paradigm for CLP referred to as Disentangled Frequency-Spatial-Temporal Attention(D-FaST). Specifically, we present an novel cognitive signal decoder that operates on disentangled frequency-space-time domain attention. This decoder encompasses three key components: frequency domain feature extraction employing multi-view attention, spatial domain feature extraction utilizing dynamic brain connection graph attention, and temporal feature extraction relying on local time sliding window attention. These components are integrated within a novel disentangled framework. Additionally, to encourage advancements in this field, we have created a new CLP dataset, MNRED. Subsequently, we conducted an extensive series of experiments, evaluating D-FaST’s performance on MNRED, as well as on publicly available datasets including ZuCo, BCIC IV-2A, and BCIC IV-2B. Our experimental results demonstrate that D-FaST outperforms existing methods significantly on both our datasets and traditional CSD datasets including establishing a state-of-the-art accuracy score 78.72% on MNRED, pushing the accuracy score on ZuCo to 78.35%, accuracy score on BCIC IV-2A to 74.85% and accuracy score on BCIC IV-2B to 76.81%.
25 March 2021 in IEEE Journal of Selected Topics in Quantum Electronics
Quantitative spectral analysis is highly demanded in many applications such as pharmaceutical research, biological sample analysis, and food quality control, in which Raman spectroscopy has been established as a label-free analytical tool. In an attempt to address the challenge faced by the classical least-squares method during the quantification of analyte concentrations, which is induced by potential inaccuracy or variations in the spectral library, we propose a modified least-squares method that accounts for such variations as individual Gaussian functions with variable parameters. Both the spectral library and the analyte concentrations are updated during iterations. The method is validated on both simulated data with multiple Gaussian peaks and experimental measurements from phantoms of chemical mixtures and K562 leukemia cells.