DSCnet: detection of drug and alcohol addiction mechanisms based on multi-angle feature learning from the hybrid representation of EEG

IntroductionDrug and alcohol addiction impair neurotransmitter systems, leading to severe physiological, psychological, and social issues. Electroencephalography (EEG) is commonly used to analyze addiction mechanisms, but traditional feature extraction methods such as time-frequency analysis, Princi...

Full description

Saved in:
Bibliographic Details
Main Authors: Jing Wu, Nan Zhang, Qilei Ye, Xiaorui Zheng, Minmin Shao, Xian Chen, Hui Huang
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-06-01
Series:Frontiers in Neuroscience
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fnins.2025.1607248/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849688494057717760
author Jing Wu
Nan Zhang
Qilei Ye
Xiaorui Zheng
Minmin Shao
Xian Chen
Hui Huang
author_facet Jing Wu
Nan Zhang
Qilei Ye
Xiaorui Zheng
Minmin Shao
Xian Chen
Hui Huang
author_sort Jing Wu
collection DOAJ
description IntroductionDrug and alcohol addiction impair neurotransmitter systems, leading to severe physiological, psychological, and social issues. Electroencephalography (EEG) is commonly used to analyze addiction mechanisms, but traditional feature extraction methods such as time-frequency analysis, Principal Component Analysis (PCA), and Independent Component Analysis (ICA) fail to capture complex relationships between variables.MethodsThis paper proposes DSCnet, a novel neural network model for addiction detection. DSCnet combines embedding layers, skip connections, depthwise separable convolution, and our self-designed Directional Adaptive Feature Modulation (DAFM) module. DAFM is a key innovation that adaptively adjusts feature directionality, extracting global features from EEG signals while preserving spatiotemporal information. This enables the model to capture neural activity patterns related to addiction mechanisms. DSCnet uses a multi-angle feature extraction strategy, emphasizing information from various perspectives.ResultsOn the drug addiction dataset, DSCnet achieved 85.11% accuracy, 85.13% precision, 85.12% recall, and 85.12% F1-score. On the UCI alcohol addiction dataset, it achieved 84.56% accuracy, 84.73% precision, 84.56% recall, and 84.63% F1-score.DiscussionThese results outperform existing models and demonstrate a balanced performance across both datasets, highlighting DSCnet's potential in addiction detection.
format Article
id doaj-art-9b77c88b8ce940a2a6b7935bc6772fd3
institution DOAJ
issn 1662-453X
language English
publishDate 2025-06-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Neuroscience
spelling doaj-art-9b77c88b8ce940a2a6b7935bc6772fd32025-08-20T03:21:59ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2025-06-011910.3389/fnins.2025.16072481607248DSCnet: detection of drug and alcohol addiction mechanisms based on multi-angle feature learning from the hybrid representation of EEGJing Wu0Nan Zhang1Qilei Ye2Xiaorui Zheng3Minmin Shao4Xian Chen5Hui Huang6College of Computer Science and Artificial Intelligence, Wenzhou University, Wenzhou, ChinaCollege of Computer Science and Artificial Intelligence, Wenzhou University, Wenzhou, ChinaData Resources Division, Wenzhou Data Bureau, Wenzhou, ChinaDepartment of Drug Rehabilitation and Correction, Wenzhou City Huanglong Compulsory Isolation Drug Rehabilitation Center, Wenzhou, ChinaDepartment of Otolaryngology, Wenzhou Central Hospital, Wenzhou, ChinaInformation Technology Center, Wenzhou Polytechnic, Wenzhou, ChinaCollege of Computer Science and Artificial Intelligence, Wenzhou University, Wenzhou, ChinaIntroductionDrug and alcohol addiction impair neurotransmitter systems, leading to severe physiological, psychological, and social issues. Electroencephalography (EEG) is commonly used to analyze addiction mechanisms, but traditional feature extraction methods such as time-frequency analysis, Principal Component Analysis (PCA), and Independent Component Analysis (ICA) fail to capture complex relationships between variables.MethodsThis paper proposes DSCnet, a novel neural network model for addiction detection. DSCnet combines embedding layers, skip connections, depthwise separable convolution, and our self-designed Directional Adaptive Feature Modulation (DAFM) module. DAFM is a key innovation that adaptively adjusts feature directionality, extracting global features from EEG signals while preserving spatiotemporal information. This enables the model to capture neural activity patterns related to addiction mechanisms. DSCnet uses a multi-angle feature extraction strategy, emphasizing information from various perspectives.ResultsOn the drug addiction dataset, DSCnet achieved 85.11% accuracy, 85.13% precision, 85.12% recall, and 85.12% F1-score. On the UCI alcohol addiction dataset, it achieved 84.56% accuracy, 84.73% precision, 84.56% recall, and 84.63% F1-score.DiscussionThese results outperform existing models and demonstrate a balanced performance across both datasets, highlighting DSCnet's potential in addiction detection.https://www.frontiersin.org/articles/10.3389/fnins.2025.1607248/fullelectroencephalogramsalcoholismdrug addictioncomputer-aided diagnosisconvolutional neural networksclassification
spellingShingle Jing Wu
Nan Zhang
Qilei Ye
Xiaorui Zheng
Minmin Shao
Xian Chen
Hui Huang
DSCnet: detection of drug and alcohol addiction mechanisms based on multi-angle feature learning from the hybrid representation of EEG
Frontiers in Neuroscience
electroencephalograms
alcoholism
drug addiction
computer-aided diagnosis
convolutional neural networks
classification
title DSCnet: detection of drug and alcohol addiction mechanisms based on multi-angle feature learning from the hybrid representation of EEG
title_full DSCnet: detection of drug and alcohol addiction mechanisms based on multi-angle feature learning from the hybrid representation of EEG
title_fullStr DSCnet: detection of drug and alcohol addiction mechanisms based on multi-angle feature learning from the hybrid representation of EEG
title_full_unstemmed DSCnet: detection of drug and alcohol addiction mechanisms based on multi-angle feature learning from the hybrid representation of EEG
title_short DSCnet: detection of drug and alcohol addiction mechanisms based on multi-angle feature learning from the hybrid representation of EEG
title_sort dscnet detection of drug and alcohol addiction mechanisms based on multi angle feature learning from the hybrid representation of eeg
topic electroencephalograms
alcoholism
drug addiction
computer-aided diagnosis
convolutional neural networks
classification
url https://www.frontiersin.org/articles/10.3389/fnins.2025.1607248/full
work_keys_str_mv AT jingwu dscnetdetectionofdrugandalcoholaddictionmechanismsbasedonmultianglefeaturelearningfromthehybridrepresentationofeeg
AT nanzhang dscnetdetectionofdrugandalcoholaddictionmechanismsbasedonmultianglefeaturelearningfromthehybridrepresentationofeeg
AT qileiye dscnetdetectionofdrugandalcoholaddictionmechanismsbasedonmultianglefeaturelearningfromthehybridrepresentationofeeg
AT xiaoruizheng dscnetdetectionofdrugandalcoholaddictionmechanismsbasedonmultianglefeaturelearningfromthehybridrepresentationofeeg
AT minminshao dscnetdetectionofdrugandalcoholaddictionmechanismsbasedonmultianglefeaturelearningfromthehybridrepresentationofeeg
AT xianchen dscnetdetectionofdrugandalcoholaddictionmechanismsbasedonmultianglefeaturelearningfromthehybridrepresentationofeeg
AT huihuang dscnetdetectionofdrugandalcoholaddictionmechanismsbasedonmultianglefeaturelearningfromthehybridrepresentationofeeg