TEDNet: Cascaded CNN-transformer with dual attentions for taste EEG decoding

IF 2.3 4区医学 Q2 BIOCHEMICAL RESEARCH METHODS

Journal of Neuroscience Methods Pub Date : 2025-09-27 DOI:10.1016/j.jneumeth.2025.110594

Xueli Wang, Guoce Lv

{"title":"TEDNet: Cascaded CNN-transformer with dual attentions for taste EEG decoding","authors":"Xueli Wang, Guoce Lv","doi":"10.1016/j.jneumeth.2025.110594","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>Traditional taste evaluation methods suffer from subjective biases and limited sensor capabilities, while existing Electroencephalogram (EEG) approaches struggle to decode complex neural patterns evoked by sour, sweet, bitter, and salty stimuli due to noise sensitivity and inadequate multi-scale feature integration.</div></div><div><h3>New method</h3><div>To address this, we propose Taste EEG Decoding Network (TEDNet), a novel deep learning architecture integrating: 1) a Temporal Spatial Convolution Module (TSCM) capturing electrode-wise dependencies, 2) a Temporal Spatial Attention Module (TSAM) adaptively reweighting critical features, and 3) a Local Global Fusion Module (LGFM) combines the local features of taste EEG with the global ones.</div></div><div><h3>Results</h3><div>Evaluated on a well-controlled dataset containing 2400 EEG samples from 30 subjects, the accuracy of TEDNet is 98.92 %, the F1-score is 98.75 %, and the Kappa coefficient is 98.49 %.</div></div><div><h3>Comparison with existing methods</h3><div>While maintaining computational efficiency, TEDNet has surpassed the existing advanced convolution and self-attention methods.</div></div><div><h3>Conclusions</h3><div>This framework establishes a robust solution for objective taste perception decoding, advancing sensory evaluation in food science.</div></div>","PeriodicalId":16415,"journal":{"name":"Journal of Neuroscience Methods","volume":"424 ","pages":"Article 110594"},"PeriodicalIF":2.3000,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Neuroscience Methods","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0165027025002389","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"BIOCHEMICAL RESEARCH METHODS","Score":null,"Total":0}

引用次数: 0

Abstract

Background

Traditional taste evaluation methods suffer from subjective biases and limited sensor capabilities, while existing Electroencephalogram (EEG) approaches struggle to decode complex neural patterns evoked by sour, sweet, bitter, and salty stimuli due to noise sensitivity and inadequate multi-scale feature integration.

New method

To address this, we propose Taste EEG Decoding Network (TEDNet), a novel deep learning architecture integrating: 1) a Temporal Spatial Convolution Module (TSCM) capturing electrode-wise dependencies, 2) a Temporal Spatial Attention Module (TSAM) adaptively reweighting critical features, and 3) a Local Global Fusion Module (LGFM) combines the local features of taste EEG with the global ones.

Results

Evaluated on a well-controlled dataset containing 2400 EEG samples from 30 subjects, the accuracy of TEDNet is 98.92 %, the F1-score is 98.75 %, and the Kappa coefficient is 98.49 %.

Comparison with existing methods

While maintaining computational efficiency, TEDNet has surpassed the existing advanced convolution and self-attention methods.

Conclusions

This framework establishes a robust solution for objective taste perception decoding, advancing sensory evaluation in food science.

查看原文本刊更多论文

具有双重关注的级联CNN-Transformer用于味觉脑电解码。

背景：传统的味觉评价方法存在主观偏差和传感器能力的限制，而现有的脑电图（EEG）方法由于噪声敏感性和多尺度特征整合不足，难以解码由酸、甜、苦和咸刺激引起的复杂神经模式。为了解决这一问题，我们提出了味觉EEG解码网络（TEDNet），这是一种新的深度学习架构，它集成了：1)捕获电极依赖关系的时空卷积模块（TSCM）， 2)自适应重新加权关键特征的时空注意模块（TSAM），以及3)将味觉EEG的局部特征与全局特征相结合的局部全局融合模块（LGFM）。结果：在控制良好的数据集（包含30名受试者的2400份EEG样本）上，TEDNet的准确率为98.92%，f1得分为98.75%，Kappa系数为98.49%。与现有方法的比较：在保持计算效率的同时，TEDNet超越了现有先进的卷积和自关注方法。结论：该框架为客观味觉解码提供了可靠的解决方案，推动了食品科学感官评价的发展。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal of Neuroscience Methods 医学-神经科学

CiteScore

7.10

自引率

3.30%

发文量

226

审稿时长

52 days

期刊介绍： The Journal of Neuroscience Methods publishes papers that describe new methods that are specifically for neuroscience research conducted in invertebrates, vertebrates or in man. Major methodological improvements or important refinements of established neuroscience methods are also considered for publication. The Journal''s Scope includes all aspects of contemporary neuroscience research, including anatomical, behavioural, biochemical, cellular, computational, molecular, invasive and non-invasive imaging, optogenetic, and physiological research investigations.