DiffMDD: A Diffusion-Based Deep Learning Framework for MDD Diagnosis Using EEG

IEEE Transactions on Neural Systems and Rehabilitation Engineering, Год журнала: 2024, Номер 32, С. 728 - 738

Опубликована: Янв. 1, 2024

Major Depression Disorder (MDD) is a common yet destructive mental disorder that affects millions of people worldwide. Making early and accurate diagnosis it very meaningful. Recently, EEG, non-invasive technique recording spontaneous electrical activity brains, has been widely used for MDD diagnosis. However, there are still some challenges in data quality size EEG: (1) A large amount noise inevitable during EEG collection, making difficult to extract discriminative features from raw EEG; (2) It recruit number subjects collect sufficient diverse model training. Both the cause overfitting problem, especially deep learning methods. In this paper, we propose DiffMDD, diffusion-based framework using EEG. Specifically, more noise-irrelevant improve model's robustness by designing Forward Diffusion Noisy Training Module. Then increase diversity help learn generalized Reverse Data Augmentation Finally, re-train classifier on augmented dataset We conducted comprehensive experiments test overall performance each module's effectiveness. The was validated two public datasets, achieving state-of-the-art performance.

Язык: Английский

---

A Review of Graph Theory-Based Diagnosis of Neurological Disorders Based on EEG and MRI

Neurocomputing, Год журнала: 2024, Номер 599, С. 128098 - 128098

Опубликована: Сен. 1, 2024

Язык: Английский

---

SFT-SGAT: A semi-supervised fine-tuning self-supervised graph attention network for emotion recognition and consciousness detection

Neural Networks, Год журнала: 2024, Номер 180, С. 106643 - 106643

Опубликована: Авг. 23, 2024

Язык: Английский

---

Fault diagnosis of mobile robot based on dual-graph convolutional network with prior fault knowledge

Advanced Engineering Informatics, Год журнала: 2024, Номер 62, С. 102865 - 102865

Опубликована: Окт. 1, 2024

Язык: Английский

---

M-Mdd: A Multi-Task Deep Learning Framework for Major Depressive Disorder Diagnosis Using Eeg

Опубликована: Янв. 1, 2025

Major Depressive Disorder (MDD) is a common and destructive psychiatric disorder worldwide. Traditional MDD diagnosis relies heavily on subjective observation questionnaires. Recently, non-invasive method of recording the brain's spontaneous activity called Electroencephalogram (EEG) has been useful tool diagnosis. However, there are still some challenges to be addressed: (1) The model's robustness EEG noise improved, (2) temporal, spectral spatial features need extracted fused appropriately. Learning both robust powerful for can improve overall performance, multi-task learning solution. In this paper, we propose M-MDD, deep framework using EEG. First, design Contrastive Noise Robustness Task learn noise-independent features. Then, Supervised Feature Extraction extract respectively, then effectively combine them together. Finally, above two modules share same feature space trained jointly with Multi-task Module, improving performance. Validated public datasets subject-independent cross-validation, our model achieves state-of-the-art

Язык: Английский

---

Graph convolution network-based eeg signal analysis: a review

Medical & Biological Engineering & Computing, Год журнала: 2025, Номер unknown

Опубликована: Янв. 30, 2025

Язык: Английский

---

M-MDD: A multi-task deep learning framework for major depressive disorder diagnosis using EEG

Neurocomputing, Год журнала: 2025, Номер unknown, С. 130008 - 130008

Опубликована: Март 1, 2025

---

Flexible Patched Brain Transformer model for EEG decoding

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 29, 2025

Abstract Decoding the human brain using non-invasive methods is a significant challenge. This study aims to enhance electroencephalography (EEG) decoding by developing of machine learning methods. Specifically, we propose novel, attention-based Patched Brain Transformer model achieve this goal. The exhibits flexibility regarding number EEG channels and recording duration, enabling effective pre-training across diverse datasets. We investigate effect data augmentation on training process. To gain insights into behavior, incorporate an inspection architecture. compare our with state-of-the-art models demonstrate superior performance only fraction parameters. results are achieved supervised pre-training, coupled time shifts as for multi-participant classification motor imagery

Язык: Английский

---

Advanced unsupervised learning: a comprehensive overview of multi-view clustering techniques

Artificial Intelligence Review, Год журнала: 2025, Номер 58(8)

Опубликована: Май 7, 2025

Язык: Английский

---

Effect of Dual‐Task Training on the Number of EEG Bands in Stroke Patients

Physiotherapy Research International, Год журнала: 2025, Номер 30(3)

Опубликована: Май 15, 2025

ABSTRACT Background/Objective Dual‐task training (DTT) positively impacts stroke recovery, but its effects on electroencephalography (EEG) using Fourier series analysis are under‐researched. This study aimed to evaluate the of DTT EEG in patients by analyzing different bands with fast transform (FFT). Methods Five participants unilateral ischemic completed 12 sessions 15‐min DTT, three times a week for 4 weeks. data were recorded before and after intervention, FFT was conducted. Assessments upper limb function, elbow flexor muscle tone, daily living activities also performed. Results showed reduction delta, theta, alpha, beta post‐DTT, while their correlation between measurement remained consistent. These changes somewhat reflected participants' improved clinical outcomes. Conclusion The results suggest that affects band frequencies, consistent pre‐ post‐intervention measurements. indicates could be useful tool assessing DTT's impact recovery.

Язык: Английский

---