Related Experiment Video
Updated: Sep 21, 2025
08:51
Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms
Published on: November 1, 2019
5.7K
A major depressive disorder diagnosis approach based on EEG signals using dictionary learning and functional
Reza Akbari Movahed1, Gila Pirzad Jahromi2, Shima Shahyad1
1Neuroscience Research Center, Baqiyatallah University of Medical Sciences, Tehran, Iran.
Physical and Engineering Sciences in Medicine
|May 31, 2022
Summary
This study introduces an automated framework using electroencephalogram (EEG) signals and dictionary learning for early major depressive disorder (MDD) diagnosis. The method achieved high accuracy, showing potential as a computer-aided diagnosis tool.
Area of Science:
- Neuroscience
- Computational Psychiatry
- Medical Imaging
Background:
- Major Depressive Disorder (MDD) significantly impacts patient behavior and daily life.
- Early diagnosis of MDD is crucial for effective treatment and preventing adverse outcomes.
- Subtle clinical manifestations of MDD pose challenges for timely diagnosis.
Purpose of the Study:
- To develop an automated framework for early diagnosis of MDD using electroencephalogram (EEG) signals.
- To leverage Dictionary Learning (DL) approaches and functional connectivity features for enhanced diagnostic accuracy.
- To evaluate the efficacy of DL-based classifiers, specifically LC-KSVD and CLC-KSVD, in distinguishing MDD patients from healthy controls.
Main Methods:
- Construction of a feature space using functional connectivity from EEG signals for MDD and healthy control (HC) participants.
- Application of Dictionary Learning (DL) based classification methods, including Label Consistent K-SVD (LC-KSVD) and Correlation-based Label Consistent K-SVD (CLC-KSVD).
- Validation using a public dataset of 34 MDD patients and 30 HC subjects, employing 10-fold cross-validation with 100 iterations.
Main Results:
- The LC-KSVD2 and CLC-KSVD2 methods demonstrated efficient classification of MDD and HC cases.
- The LCKSVD2 method achieved the highest classification performance, with an average accuracy of 99.0%, sensitivity of 98.9%, specificity of 99.2%, F1-score of 99.0%, and false discovery rate of 0.8%.
- The proposed framework exhibited accurate performance in differentiating between MDD patients and healthy individuals.
Conclusions:
- The developed EEG-based automated framework shows significant promise for the early diagnosis of MDD.
- The Dictionary Learning approach, particularly LC-KSVD2, is effective in analyzing functional connectivity features for MDD classification.
- The findings suggest the potential for this method to be developed into a valuable computer-aided diagnosis (CAD) tool for MDD.
