CNN-Transformer encoder-based model for depressive state recognition with multidimensional EEG feature fusion

doi:10.11772/j.issn.1001-9081.2025070908

Journal of Computer Applications ›› 2026, Vol. 46 ›› Issue (9): 3025-3034.DOI: 10.11772/j.issn.1001-9081.2025070908

• Frontier and comprehensive applications • Previous Articles

CNN-Transformer encoder-based model for depressive state recognition with multidimensional EEG feature fusion

Siyu DONG¹, Xin JIA², Mi ZHANG², Xinwen CHEN¹, Xinying SUN¹(), Fangxiao CHENG²()

^1.School of Public Health，Peking University，Beijing 100191，China
^2.Institute of Medical Technology，Peking University，Beijing 100191，China

Received:2025-08-11 Revised:2025-09-08 Accepted:2025-09-11 Online:2025-11-05 Published:2026-09-10
Contact: Xinying SUN, Fangxiao CHENG
About author:DONG Siyu， born in 2001， Ph. D. candidate. Her research interests include health big data mining， medical artificial intelligence.
JIA Xin， born in 2001， M. S. candidate. His research interests include neuroelectrophysiological signal processing and feature extraction.
ZHANG Mi， born in 2001， M. S. candidate. Her research interests include radiomics and machine learning model development.
CHEN Xinwen， born in 1996， Ph. D. candidate. His research interests include electroencephalogram data analysis.
SUN Xinying， born in 1973， Ph. D.， professor. Her research interests include artificial intelligence， mobile health interventions.
CHENG Fangxiao， born in 1993， Ph. D.， associate research fellow. His research interests include neural information acquisition and processing， detection and regulation of brain ultrastructure （extracellular space of brain cells）.
Supported by:
Beijing Natural Science Foundation-Xiaomi Innovation Joint Fund(L243037)

基于CNN-Transformer编码器和多维脑电特征融合的抑郁状态识别模型

董思语¹, 贾鑫², 张密², 陈新文¹, 孙昕霙¹(), 程方骁²()

^1.北京大学公共卫生学院，北京 100191
^2.北京大学医学技术研究院，北京 100191

通讯作者: 孙昕霙,程方骁
作者简介:董思语（2001—），女，山东济宁人，博士研究生，主要研究方向：医疗大数据挖掘、医学人工智能
贾鑫（2001—），男，河北衡水人，硕士研究生，主要研究方向：神经电生理信号处理与特征提取
张密（2001—），女，重庆人，硕士研究生，主要研究方向：影像组学与机器学习模型开发
陈新文（1996—），男，江西南昌人，博士研究生，主要研究方向：脑电数据分析
孙昕霙（1973—），女，吉林吉林人，教授，博士，主要研究方向：人工智能、移动健康干预
程方骁（1993—），男，辽宁抚顺人，副研究员，博士，主要研究方向：神经信息获取与处理、脑超微结构（脑细胞外间隙）的探测与调控。
基金资助:
北京市自然科学基金-小米创新联合基金资助项目(L243037)

Abstract

Abstract:

Considering the challenges of insufficient feature fusion and limited classification granularity in ElectroEncephaloGram （EEG） -based depression recognition task， a CNN-Transformer Encoder-based model for Depressive State Recognition （CTE-DSR） that integrated multidimensional EEG features was proposed. First， One-Dimensional Convolutional Neural Network （1D-CNN） was used to extract local spatial sequential features from EEG. Then， Transformer encoder module was introduced to model global correlations between leads. Simultaneously， Multi-Layer Perceptron （MLP） was used to embed depression-related non-sequential features into a high-dimensional space. Finally， the two features were fused to implement fine-grained classification of depression states through a fully connected layer. Besides， a loss weighting strategy was employed to improve the model's ability to identify minority classes. Experimental results show that compared with other five machine learning models （such as Random Forest （RF） and Adaptive Boosting （AdaBoost）） and two combination models （CNN-LSTM and CNN-RNN）， CTE-DSR achieves at least 23.0% and 6.7% higher recall and Area Under Curve （AUC） in binary classification， respectively， and at least 15.6% and 1.0% higher F1-score and macro-average AUC in ternary classification， respectively. It can be seen that the proposed model has performance advantages in identifying depressive states and can be used to assist in early screening and relapse risk assessment of depression.

Key words: ElectroEncephaloGram (EEG) signal, depression multi-state recognition, Convolutional Neural Network (CNN), Transformer encoder, feature fusion, neural biomarker

摘要：

针对脑电（EEG）信号在识别抑郁症任务中存在的特征融合不足和分类粒度有限等问题，提出一种基于CNN-Transformer编码器的多维脑电特征融合抑郁状态识别模型（CTE-DSR）。首先，利用一维卷积神经网络（1D-CNN）提取EEG的局部空间序列特征；其次，引入Transformer编码器建模各导联间的全局相关性，同时，通过多层感知机（MLP）将抑郁相关的非序列特征嵌入高维空间；最后，将2类特征融合并通过全连接层实现对抑郁状态的细粒度分类；此外，采用损失加权策略改善模型对少数类的识别能力。实验结果表明，与其他5类机器学习模型（如随机森林（RF）和自适应提升算法（AdaBoost）等）和2类组合模型（CNN-LSTM和CNN-RNN）相比， CTE-DSR在二分类任务中的召回率和曲线下面积（AUC）分别至少提升了23.0%和6.7%，在三分类中的F1值和宏平均AUC分别至少提升了15.6%和1.0%。可见，所提模型在抑郁状态识别中展现出性能优势，可用于辅助抑郁症早期筛查和复发风险评估。

关键词: 脑电信号, 抑郁症多状态识别, 卷积神经网络, Transformer编码器, 特征融合, 神经生物标志物

CLC Number:

TP391.4

Siyu DONG, Xin JIA, Mi ZHANG, Xinwen CHEN, Xinying SUN, Fangxiao CHENG. CNN-Transformer encoder-based model for depressive state recognition with multidimensional EEG feature fusion[J]. Journal of Computer Applications, 2026, 46(9): 3025-3034.

董思语, 贾鑫, 张密, 陈新文, 孙昕霙, 程方骁. 基于CNN-Transformer编码器和多维脑电特征融合的抑郁状态识别模型[J]. 《计算机应用》唯一官方网站, 2026, 46(9): 3025-3034.

Figures/Tables 7

References 47

[1]	吴炳鑫，周易，韩思博，等. “疏肝调神”针刺对脂多糖诱导抑郁模型小鼠海马TLR4/NF-κB通路的影响［J］.中医杂志， 2025， 66（5）： 526-531.
	Wu Bingxin， Zhou Yi， Han Sibo， et al. Effect of acupuncture with the method of soothing the liver and regulating the mind on the TLR4/NF-κB pathway in the hippocampus of lipopolysaccharide-induced depression model mice ［J］. Journal of Traditional Chinese Medicine， 2025， 66（5）： 526-531.
[2]	Liu Z， Kuai M. The global burden of depression in adolescents and young adults， 1990-2021： systematic analysis of the global burden of disease study ［J］. BMC Psychiatry， 2025， 25： No.767.
[3]	傅小兰，张侃，陈雪峰，等.心理健康蓝皮书：中国国民心理健康发展报告（2021~2022）［M］.北京：社会科学文献出版社， 2023.
	Fu Xiaolan， Zhang Kan， Chen Xuefeng， et al. Blue book on mental health： report on national mental health development in China （2021-2022）［M］. Beijing： Social Sciences Academic Press （China）， 2023.
[4]	Zimmerman M， Martinez J H， Young D， et al. Severity classification on the Hamilton depression rating scale ［J］. Journal of Affective Disorders， 2013， 150（2）： 384-388.
[5]	Katz I R， Liebmann E P， Resnick S G， et al. Performance of the PHQ-9 across conditions and comorbidities： findings from the Veterans Outcome Assessment survey ［J］. Journal of Affective Disorders， 2021， 294： 864-867.
[6]	Zigler C K， Coles T. Clinical outcome assessments are never “validated” ［J］. Value in Health， 2024， 27（11）： 1494-1496.
[7]	黄恋.基于多任务情绪刺激声学特征的抑郁倾向识别［D］.贵阳：贵州师范大学， 2025： 2-3.
	Huang Lian. Depression tendency identification based on multitask emotional stimulus acoustic features ［D］. Guiyang： Guizhou Normal University， 2025： 2-3.
[8]	Acharya U R， Oh S L， Hagiwara Y， et al. Automated EEG-based screening of depression using deep convolutional neural network ［J］. Computer Methods and Programs in Biomedicine， 2018， 161： 103-113.
[9]	Ksibi A， Zakariah M， Menzli L J， et al. Electroencephalography-based depression detection using multiple machine learning techniques ［J］. Diagnostics， 2023， 13（10）： No.1779.
[10]	Badr Y， Tariq U， Al-Shargie F， et al. A review on evaluating mental stress by deep learning using EEG signals ［J］. Neural Computing and Applications， 2024， 36（21）： 12629-12654.
[11]	Kang M， Kang S， Lee Y. Prediction of Beck depression inventory score in EEG： application of deep-asymmetry method ［J］. Applied Sciences， 2021， 11（19）： No.9218.
[12]	Hosseinifard B， Moradi M H， Rostami R. Classifying depression patients and normal subjects using machine learning techniques and nonlinear features from EEG signal ［J］. Computer Methods and Programs in Biomedicine， 2013， 109（3）： 339-345.
[13]	Elnaggar K， El-Gayar M M， Elmogy M. Depression detection and diagnosis based on ElectroencEphaloGram （EEG） analysis： a systematic review ［J］. Diagnostics， 2025， 15（2）： No.210.
[14]	Liu W， Zheng W L， Lu B L. Emotion recognition using multimodal deep learning ［C］// ICONIP 2016. Cham： Springer， 2016： 521-529.
[15]	Chang J， Choi Y. Depression diagnosis based on electroencephalography power ratios ［J］. Brain and Behavior， 2023， 13（8）： No.e3173.
[16]	Boby K， Veerasingam S. Depression diagnosis： EEG-based cognitive biomarkers and machine learning ［J］. Behavioural Brain Research， 2025， 478： No.115325.
[17]	Huang W， Chang W， Yan G， et al. Spatio-spectral feature classification combining 3D-convolutional neural networks with long short-term memory for motor movement/imagery ［J］. Engineering Applications of Artificial Intelligence， 2023， 120： No.105862.
[18]	Prabhakar S K， Lee S W. Improved sparse representation based robust hybrid feature extraction models with transfer and deep learning for EEG classification ［J］. Expert Systems with Applications， 2022， 198： No.116783.
[19]	Monroe S M， Harkness K L. Major depression and its recurrences： life course matters ［J］. Annual Review of Clinical Psychology， 2022， 18： 329-357.
[20]	Lee P F， Kan D P X， Croarkin P， et al. Neurophysiological correlates of depressive symptoms in young adults： a quantitative EEG study ［J］. Journal of Clinical Neuroscience， 2018， 47： 315-322.
[21]	Movahed R A， Jahromi G P， Shahyad S， et al. A major depressive disorder diagnosis approach based on EEG signals using dictionary learning and functional connectivity features ［J］. Physical and Engineering Sciences in Medicine， 2022， 45（3）： 705-719.
[22]	Khadidos A O， Alyoubi K H， Mahato S， et al. Machine learning and electroencephalogram signal based diagnosis of depression ［J］. Neuroscience Letters， 2023， 809： No.137313.
[23]	Lee K S， Ham B J. Machine learning on early diagnosis of depression ［J］. Psychiatry Investigation， 2022， 19（8）： 597-605.
[24]	Zhu T， Luo W， Yu F. Convolution-and attention-based neural network for automated sleep stage classification ［J］. International Journal of Environmental Research and Public Health， 2020， 17（11）： No.4152.
[25]	Ke H， Cai C， Wang F， et al. Interpretation of frequency channel-based CNN on depression identification ［J］. Frontiers in Computational Neuroscience， 2021， 15： No.773147.
[26]	Zheng X， Chen W. An attention-based bi-LSTM method for visual object classification via EEG ［J］. Biomedical Signal Processing and Control， 2021， 63： No.102174.
[27]	Masuda N， Yairi I E. Multi-input CNN-LSTM deep learning model for fear level classification based on EEG and peripheral physiological signals ［J］. Frontiers in Psychology， 2023， 14： No.1141801.
[28]	Zeynali M， Seyedarabi H， Afrouzian R. Classification of EEG signals using Transformer based deep learning and ensemble models ［J］. Biomedical Signal Processing and Control， 2023， 86（Pt A）： No.105130.
[29]	Zhu J， Li Y， Yang C， et al. Transformer-based fusion model for mild depression recognition with EEG and pupil area signals ［J］. Medical and Biological Engineering and Computing， 2025， 63（7）： 2011-2027.
[30]	Xie W， Wang C， Lin Z， et al. Multimodal fusion diagnosis of depression and anxiety based on CNN-LSTM model ［J］. Computerized Medical Imaging and Graphics， 2022， 102： No.102128.
[31]	Zhang W， Mao K， Chen J. A multimodal approach for detection and assessment of depression using text， audio and video ［J］. Phenomics， 2024， 4（3）： 234-249.
[32]	Zhou Y， Han W， Yao X， et al. Developing a machine learning model for detecting depression， anxiety， and apathy in older adults with mild cognitive impairment using speech and facial expressions： a cross-sectional observational study ［J］. International Journal of Nursing Studies， 2023， 146： No.104562.
[33]	Tigga N P， Garg S. Efficacy of novel attention-based gated recurrent units Transformer for depression detection using electroencephalogram signals ［J］. Health Information Science and Systems， 2022， 11： No.1.
[34]	Rawat K， Sharma T. An enhanced CNN-Bi-transformer based framework for detection of neurological illnesses through neurocardiac data fusion ［J］. Scientific Reports， 2025， 15： No.11379.
[35]	Maggi G， D'Iorio A， Aiello E N， et al. Psychometrics and diagnostics of the Italian version of the Beck Depression Inventory-Ⅱ （BDI-Ⅱ） in Parkinson's disease ［J］. Neurological Sciences， 2023， 44（5）： 1607-1612.
[36]	Mumtaz W， Qayyum A. A deep learning framework for automatic diagnosis of unipolar depression ［J］. International Journal of Medical Informatics， 2019， 132： No.103983.
[37]	Shivcharan M， Boby K， Sridevi V. EEG based machine learning models for automated depression detection ［C］// IEEE CONECCT 2023. Piscataway： IEEE， 2023： 1-6.
[38]	Cavanagh J F. EEG： depression rest ［DS/OL］. （2021-01-18）［2025-08-19］. .
[39]	Cavanagh J F， Bismark A W， Frank M J， et al. Multiple dissociations between comorbid depression and anxiety on reward and punishment processing： evidence from computationally informed EEG ［J］. Computational Psychiatry， 2019， 3： 1-17.
[40]	Zhang X S， Roy R J， Jensen E W. EEG complexity as a measure of depth of anesthesia for patients ［J］. IEEE Transactions on Biomedical Engineering， 2001， 48（12）： 1424-1433.
[41]	Richman J S， Moorman J R. Physiological time-series analysis using approximate entropy and sample entropy ［J］. American Journal of Physiology-Heart and Circulatory Physiology， 2000， 278（6）： H2039-H2049.
[42]	Bandt C， Pompe B. Permutation entropy： a natural complexity measure for time series ［J］. Physical Review Letters， 2002， 88（17）： No.174102.
[43]	Younesi A， Ansari M， Fazli M， et al. A comprehensive survey of convolutions in deep learning： applications， challenges， and future trends ［J］. IEEE Access， 2024， 12： 41180-41218.
[44]	Li Y， Shen Y， Fan X， et al. A novel EEG-based major depressive disorder detection framework with two-stage feature selection ［J］. BMC Medical Informatics and Decision Making， 2022， 22： No.209.
[45]	Roh S C， Kim J S， Kim S， et al. Frontal alpha asymmetry moderated by suicidal ideation in patients with major depressive disorder： a comparison with healthy individuals ［J］. Clinical Psychopharmacology and Neuroscience， 2020， 18（1）： 58-66.
[46]	Whitton A E， Deccy S， Ironside M L， et al. Electroencephalography source functional connectivity reveals abnormal high-frequency communication among large-scale functional networks in depression ［J］. Biological Psychiatry： Cognitive Neuroscience and Neuroimaging， 2018， 3（1）： 50-58.
[47]	Benschop L， Poppa T， Medani T， et al. Electrophysiological scarring in remitted depressed patients： elevated EEG functional connectivity between the posterior cingulate cortex and the subgenual prefrontal cortex as a neural marker for rumination ［J］. Journal of Affective Disorders， 2021， 281： 493-501.

模型	二分类					三分类
模型	Acc	Prec	Rec	F1	AUC	Acc	Prec	Rec	F1	Mac_AUC	Mic_AUC	Wt_AUC
CTE-DSR	0.761	0.615	0.727	0.667	0.775	0.831	0.814	0.604	0.665	0.760	0.900	0.739
SVM	0.687	0.522	0.545	0.533	0.702	0.763	0.254	0.333	0.288	0.741	0.891	0.727
RF	0.582	0.406	0.591	0.481	0.674	0.780	0.592	0.381	0.375	0.728	0.891	0.726
AdaBoost	0.522	0.361	0.591	0.448	0.509	0.678	0.538	0.618	0.562	0.752	0.849	0.763
XGB	0.642	0.450	0.409	0.429	0.692	0.763	0.483	0.414	0.420	0.721	0.888	0.734
LGB	0.672	0.500	0.545	0.522	0.721	0.712	0.364	0.351	0.343	0.705	0.884	0.706
CNN-RNN	0.657	0.481	0.591	0.531	0.671	0.746	0.585	0.567	0.575	0.677	0.840	0.681
CNN-LSTM	0.731	0.643	0.409	0.500	0.726	0.712	0.486	0.472	0.463	0.739	0.867	0.736

模型	二分类					三分类
模型	Acc	Prec	Rec	F1	AUC	Acc	Prec	Rec	F1	Mac_AUC	Mic_AUC	Wt_AUC
CTE-DSR	0.761	0.615	0.727	0.667	0.775	0.831	0.814	0.604	0.665	0.760	0.900	0.739
SVM	0.687	0.522	0.545	0.533	0.702	0.763	0.254	0.333	0.288	0.741	0.891	0.727
RF	0.582	0.406	0.591	0.481	0.674	0.780	0.592	0.381	0.375	0.728	0.891	0.726
AdaBoost	0.522	0.361	0.591	0.448	0.509	0.678	0.538	0.618	0.562	0.752	0.849	0.763
XGB	0.642	0.450	0.409	0.429	0.692	0.763	0.483	0.414	0.420	0.721	0.888	0.734
LGB	0.672	0.500	0.545	0.522	0.721	0.712	0.364	0.351	0.343	0.705	0.884	0.706
CNN-RNN	0.657	0.481	0.591	0.531	0.671	0.746	0.585	0.567	0.575	0.677	0.840	0.681
CNN-LSTM	0.731	0.643	0.409	0.500	0.726	0.712	0.486	0.472	0.463	0.739	0.867	0.736

模型	平衡化方式	二分类					三分类
模型	平衡化方式	Acc	Prec	Rec	F1	AUC	Acc	Prec	Rec	F1	Mac_AUC	Mic_AUC	Wt_AUC
CTE-DSR	None	0.687	0.667	0.091	0.160	0.552	0.797	0.692	0.589	0.615	0.756	0.900	0.743
	SMOTE	0.687	0.524	0.500	0.512	0.707	0.780	0.636	0.502	0.529	0.791	0.915	0.794
	Weighted	0.761	0.615	0.727	0.667	0.775	0.831	0.814	0.604	0.665	0.760	0.900	0.739
XGB	None	0.687	0.533	0.364	0.432	0.550	0.763	0.429	0.374	0.365	0.739	0.894	0.768
	SMOTE	0.672	0.500	0.500	0.500	0.674	0.746	0.518	0.447	0.467	0.722	0.878	0.705
	Weighted	0.642	0.450	0.409	0.429	0.692	0.763	0.483	0.414	0.420	0.721	0.888	0.734
RNN	None	0.672	0.500	0.136	0.214	0.668	0.724	0.374	0.359	0.356	0.704	0.875	0.708
	SMOTE	0.657	0.481	0.591	0.531	0.671	0.712	0.551	0.512	0.527	0.727	0.848	0.697
	Weighted	0.642	0.450	0.409	0.429	0.600	0.746	0.585	0.567	0.575	0.677	0.840	0.681

模型	平衡化方式	二分类					三分类
模型	平衡化方式	Acc	Prec	Rec	F1	AUC	Acc	Prec	Rec	F1	Mac_AUC	Mic_AUC	Wt_AUC
CTE-DSR	None	0.687	0.667	0.091	0.160	0.552	0.797	0.692	0.589	0.615	0.756	0.900	0.743
	SMOTE	0.687	0.524	0.500	0.512	0.707	0.780	0.636	0.502	0.529	0.791	0.915	0.794
	Weighted	0.761	0.615	0.727	0.667	0.775	0.831	0.814	0.604	0.665	0.760	0.900	0.739
XGB	None	0.687	0.533	0.364	0.432	0.550	0.763	0.429	0.374	0.365	0.739	0.894	0.768
	SMOTE	0.672	0.500	0.500	0.500	0.674	0.746	0.518	0.447	0.467	0.722	0.878	0.705
	Weighted	0.642	0.450	0.409	0.429	0.692	0.763	0.483	0.414	0.420	0.721	0.888	0.734
RNN	None	0.672	0.500	0.136	0.214	0.668	0.724	0.374	0.359	0.356	0.704	0.875	0.708
	SMOTE	0.657	0.481	0.591	0.531	0.671	0.712	0.551	0.512	0.527	0.727	0.848	0.697
	Weighted	0.642	0.450	0.409	0.429	0.600	0.746	0.585	0.567	0.575	0.677	0.840	0.681

模型	Acc	Prec	Rec	F1	Mac_AUC	Mic_AUC	Wt_AUC
CTE-DSR	0.831	0.814	0.604	0.665	0.760	0.900	0.739
-CNN	0.627	0.360	0.395	0.369	0.614	0.791	0.629
-Transformer	0.644	0.468	0.523	0.483	0.763	0.857	0.718
-Positional Encoding	0.763	0.592	0.575	0.582	0.669	0.855	0.674
-Lowdim MLP	0.695	0.458	0.424	0.428	0.616	0.837	0.604

CNN-Transformer encoder-based model for depressive state recognition with multidimensional EEG feature fusion

基于CNN-Transformer编码器和多维脑电特征融合的抑郁状态识别模型

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 7

References 47

Related Articles 15

Recommended Articles

Metrics

[1]	Dingli YANG, Yuanfang WEI, Wenrui HU, Liyang KONG, Yinshan YU. Spatio-temporal modeling and hierarchical feature enhancement for person re-identification [J]. Journal of Computer Applications, 2026, 46(9): 2776-2786.
[2]	Kan HE, Hongfeng MA, Xuejun ZONG, Hongyan SHI, Lian LIAN, Bowei NING. DCSFFuzzer： dual-channel semantic feature fusion-based fuzz testing method for industrial control protocols [J]. Journal of Computer Applications, 2026, 46(9): 2877-2888.
[3]	Yixin ZHANG, Lin JIANG, Yuancheng LI, Chen JI. CNN pruning and quantization compression method for reconfigurable structures [J]. Journal of Computer Applications, 2026, 46(9): 2732-2740.
[4]	Jiangyan CHEN, Yandan WANG, Yihu LIU, Yinglong MA. Collaborative query optimization-based framework for long-tailed object detection [J]. Journal of Computer Applications, 2026, 46(9): 2769-2775.
[5]	Lin HUANG, Yingshun LI, Weiyan TONG, Shuyuan ZHANG, Zihan WANG. Remote sensing image object detection algorithm based on improved real-time detection Transformer [J]. Journal of Computer Applications, 2026, 46(8): 2652-2659.
[6]	Ming LIU, Dongqi SHEN, Ziyang MENG. Point cloud registration network with dual-branch multi-level feature fusion [J]. Journal of Computer Applications, 2026, 46(8): 2584-2593.
[7]	Jiangyu YAN, Shiqian WANG, Hong WANG. Electricity theft detection method based on improved Wasserstein generative adversarial network with gradient penalty and time-frequency feature fusion [J]. Journal of Computer Applications, 2026, 46(8): 2691-2698.
[8]	Guoyou ZHANG, Hongyu NIE, Lihu PAN, Rundong LEI. Point-MLPBLS： point cloud semantic segmentation network based on MLP cascaded broad learning system [J]. Journal of Computer Applications, 2026, 46(7): 2259-2266.
[9]	Chao LYU, Geyao MA. Lightweight human pose estimation network based on redundant feature suppression [J]. Journal of Computer Applications, 2026, 46(6): 1973-1980.
[10]	Songhao ZHU, Zhiyun ZHAO, Mengling WANG. Low-light image enhancement network based on lightweight residual and brightness-aware dynamic feature fusion [J]. Journal of Computer Applications, 2026, 46(6): 1936-1946.
[11]	Xuechao LIAO, Rui CHEN. Prediction-evaluation framework for anomaly detection in electric vehicle lithium-ion battery [J]. Journal of Computer Applications, 2026, 46(5): 1614-1623.
[12]	Hongrui ZHANG, Weiming FENG, Luxia YANG, Yongjie MA. CSAF-YOLO： improved YOLO11 algorithm for underwater small object detection [J]. Journal of Computer Applications, 2026, 46(5): 1578-1585.
[13]	Yuqian HUANG, Hui HUANG, Yongbin QIN, Ruizhang HUANG, Yanping CHEN, Yulin ZHOU, Qian SUN. Judicial element extraction method by integrating global and local semantics [J]. Journal of Computer Applications, 2026, 46(5): 1460-1467.
[14]	Xinyao LIU, Jun LIANG, Jiahao LONG, Renliang YAN. Fine-grained Chinese herbal medicine image classification based on feature fusion and channel information compensation [J]. Journal of Computer Applications, 2026, 46(5): 1677-1683.
[15]	Jiali ZHENG, Gang ZHOU, Jing CHEN, Shunhang LI. Adaptive multi-feature fusion detection method for AI-generated text [J]. Journal of Computer Applications, 2026, 46(5): 1433-1440.