The team combined two complementary ways of analyzing resting-state electroencephalography (EEG). The first is EEG microstate analysis, a technique that treats the moment-to-moment topography of scalp electrical fields as a sequence of brief, quasi-stable configurations. Researchers have long identified four canonical microstate classes, labeled A through D, each thought to reflect a different large-scale brain network: microstate A has been associated with phonological and auditory processing, microstate B with visual networks, microstate C with the salience and default mode networks involved in self-referential thought, and microstate D with attention and executive control. Rather than measuring static levels of activity, microstate analysis captures the dynamics of how the brain switches between these configurations over time, including parameters such as duration, occurrence, coverage, and, critically, the probabilities of transitioning from one state to another.

The second approach is source-space functional connectivity analysis. Scalp EEG electrodes record a mixture of signals from many brain regions, so the researchers used low-resolution electromagnetic tomography analysis (LORETA) to reconstruct the likely cortical sources of the electrical activity and then computed connectivity between predefined regions of interest, organized by Brodmann areas. Connectivity was quantified with the phase lag index (PLI), a measure of the consistency of phase relationships between signals that is relatively robust to the volume conduction artifacts that can otherwise inflate apparent connectivity in EEG data. The idea was that depression, increasingly understood as a disorder of brain network organization rather than a single localized defect, might leave detectable traces in how these reconstructed regions communicate with one another at rest.

To test these ideas, the researchers collected resting-state EEG from 115 patients diagnosed with major depressive disorder and 43 healthy controls. The clinical cohort was thoroughly characterized: participants were assessed with instruments including the 21-item Hamilton Depression Rating Scale (HAMD-21), the Hamilton Anxiety Rating Scale, the Childhood Trauma Questionnaire, the Social Support Rate Scale, the Patient Health Questionnaire-15, the Pittsburgh Sleep Quality Index, the Chinese Perceived Stress Scale, the Temporal Experience of Pleasure Scale, and the Digit Symbol Substitution Test. The study was approved by the Ethics Committee of Tongji Hospital and prospectively registered on the Chinese Clinical Trial Registry (ChiCTR2200057365) before data collection began, and all participants provided written informed consent in accordance with the Declaration of Helsinki.

The EEG data themselves underwent a rigorous preprocessing pipeline. Signals were filtered, artifacts were removed using independent component analysis (ICA), and epochs were segmented for analysis. Microstates were identified on the basis of global field power peaks, and the group differences in microstate parameters and transition probabilities were tested statistically with correction for multiple comparisons using the false discovery rate (FDR) procedure, a standard safeguard against spurious findings when many features are tested simultaneously. This matters because EEG datasets contain hundreds or thousands of candidate features, and uncorrected analyses are notoriously prone to producing differences that vanish upon replication.

After FDR correction, one microstate finding stood firm: patients with depression showed a significantly increased transition probability from microstate D to microstate B, with a corrected p-value of 0.048. In the framework that links microstates to large-scale networks, this suggests that in depression the brain’s moment-to-moment dynamics are biased toward shifting from an attention- and executive-control-related configuration into a visual-network configuration more often than in healthy brains. The remaining microstate parameters did not survive correction, making this single altered transition the most robust electrophysiological difference between the groups in the study. It is a subtle signal, but it is precisely the kind of dynamic, network-level signature that microstate researchers have proposed as characteristic of psychiatric illness.

The source-space functional connectivity results told a more cautionary tale. At a lenient, uncorrected statistical threshold of p less than 0.05, the analysis revealed several edge-wise differences between patients and controls, hinting at altered communication among cortical regions in depression. However, when the researchers applied FDR correction across all the connectivity edges, none of these differences survived. This outcome is scientifically valuable in its own right: it demonstrates that while resting-state EEG connectivity in source space may contain group-level hints of depression-related alterations, individual edge-level effects are not yet strong or consistent enough to serve as reliable biomarkers in a cohort of this size. The authors were transparent about this limitation rather than overstating the connectivity findings.

The heart of the study, however, was its machine learning analysis. The team trained a support vector machine (SVM), a classical and well-understood classifier, to distinguish patients from controls using three different feature sets: microstate features alone, functional connectivity features alone, and a weighted fusion of the two. Classification performance was evaluated with the area under the receiver operating characteristic curve (AUC), a standard metric where 0.5 represents chance and 1.0 represents perfect discrimination. The microstate-only model achieved an AUC of 0.724 (95 percent confidence interval 0.623 to 0.818), the connectivity-only model reached 0.747 (0.642 to 0.841), and the combined, weighted fusion model performed best numerically with an AUC of 0.759 (0.660 to 0.847). A sensitivity analysis adjusting for age, gender, and body mass index was also performed to ensure the results were not confounded by demographic differences between groups.

Those numbers place the approach in a meaningful but honest context. An AUC of roughly 0.76 indicates discrimination well above chance, meaning the EEG-derived features carry genuine diagnostic information about depression, but it falls short of the performance that would be required of a stand-alone clinical diagnostic test. The confidence intervals, which span from the low 0.6s to the mid 0.8s, reflect the uncertainty inherent in a sample of 158 participants. The authors themselves are explicit that the integration of microstate and functional connectivity features may provide potentially useful EEG-derived information for machine learning classification, but that its generalizability and clinical utility require validation in larger independent cohorts before any translation into practice.

Even so, the study represents a promising step toward an objective, biology-informed approach to depression. EEG is cheap, widely available, painless, and far more accessible than functional MRI or positron emission tomography, which makes it an attractive platform for scalable biomarker development. By showing that the temporal choreography of the brain’s electrical microstates, particularly the flow from attention-related state D into visual state B, differs measurably in depression, and that combining temporal dynamics with source-space connectivity modestly improves machine learning classification, the Tongji-led team has sketched a template for future work: larger, multi-site cohorts, rigorous multiple-comparison correction, and transparent reporting of both surviving and non-surviving findings. If that template is followed, the long-standing reliance on subjective symptom reports alone may one day be supplemented by a few minutes of recorded brain activity, read not by a single number but by a pattern of the mind’s fastest rhythms.

Subject of Research: EEG microstate dynamics and source-space functional connectivity as machine learning features for classifying major depressive disorder

Article Title: EEG microstate dynamics and source-space functional connectivity features for machine learning classification of major depressive disorder

Article References: Li, C., Shi, K., Song, Y., Xia, Y., Wang, Z., Feng, J., Wang, K., Zhang, H., & Yang, Y. (2026). EEG microstate dynamics and source-space functional connectivity features for machine learning classification of major depressive disorder. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08625-6

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08625-6

Keywords: major depressive disorder, EEG, microstates, functional connectivity, machine learning, support vector machine, biomarkers, LORETA, phase lag index, psychiatry, neurophysiology, BMC Psychiatry

The foundation of the work lies in a phenomenon known as EEG microstates. When researchers record electrical activity from the scalp using electroencephalography, they observe that the spatial configuration of the brain’s electric field does not change randomly from instant to instant. Instead, it remains stable for roughly 80 to 120 milliseconds before abruptly jumping to a new configuration. These brief, quasi-stable maps of scalp potential are called microstates, and decades of research have shown that they can be grouped into a small set of canonical classes, conventionally labeled A, B, C, and D. Each class is thought to reflect the momentary dominance of a distinct large-scale brain network, linking microstates to the resting-state networks famously studied with functional MRI.

Most previous attempts to use microstates for depression research have treated them statically. Investigators typically compute summary measures such as how long each microstate lasts, how often it appears, and what fraction of total recording time it occupies, and then compare those averages between patients and healthy volunteers. Such measures have repeatedly revealed abnormalities in depression, but they discard a great deal of information. The brain does not simply cycle through microstates at random; the order in which one state follows another encodes the dynamics of how neural networks hand control back and forth. Prior clinical work has already hinted that this choreography is disturbed in MDD, with altered transition probabilities between specific microstate pairs, and with microstate abnormalities linked to cognitive dysfunction and inflammatory markers in depressed patients.

The Changzhou-led team went a step further than pairwise transitions. Rather than asking only whether state C tends to be followed by state D, they mined higher-order rules, sequences in which one microstate reliably follows another, which in turn follows a third. To do this they employed the RuleGrowth algorithm, a sequential pattern-mining method originally developed in computer science to discover rules that recur across many sequences. Applied to microstate sequences, RuleGrowth can extract rules such as C followed by D followed by B, along with statistical measures of how often the rule occurs and how confident one can be that the full sequence will complete once its beginning is observed. This turns a stream of brain states into a vocabulary of recurring phrases, some of which may be spoken differently by depressed and healthy brains.

One technical challenge immediately arises: different people produce different amounts of EEG data, and therefore different numbers of microstate transitions and mined rules. A classifier that expects a fixed-length feature vector struggles with such variable-structure input. The researchers solved this with multiple instance learning, a machine learning paradigm designed for problems where labels apply to bags of instances rather than to individual items. In their framework, each participant is treated as a bag containing all of that person’s mined transition rules, and the model learns from attention-weighted combinations of the instances within each bag. An attention mechanism allows the classifier to focus on the rules that are most diagnostic of depression while down-weighting irrelevant ones, an approach that echoes attention-based deep multiple instance learning methods developed for other domains.

The team evaluated the framework on two datasets. The first was MODMA, the multi-modal open dataset for mental disorder analysis, a publicly available resource that has become a standard benchmark for EEG-based mental health research. The second was a dataset collected by the researchers themselves. On MODMA, the rule-driven model reached an accuracy of 96.67 percent, an F1-score of 94.12 percent, and an area under the receiver operating characteristic curve of 97.23 percent. On the self-collected dataset, it achieved 92.67 percent accuracy, 92.78 percent F1-score, and 94.50 percent AUC. In both cases, the approach outperformed conventional classifiers built on static microstate features, providing direct evidence that the temporal structure of microstate sequences carries diagnostic information that average-based measures leave on the table.

Beyond classification performance, the study delivered findings of genuine neuroscientific interest. After applying false discovery rate correction to control for the many statistical comparisons involved in mining large numbers of rules, the researchers found significant group differences in higher-order transition patterns in both datasets, with seven significant rules identified in one and thirteen in the other. Crucially, two rules appeared consistently across both datasets: the sequences C to D to B and A to D to B showed higher confidence in the MDD group than in healthy controls. In other words, when the depressed brain entered certain states, it was more likely than the healthy brain to complete the same downstream sequences, suggesting that depression is associated with disrupted higher-order switching dynamics among the large-scale networks that microstates are believed to index.

The specific microstate classes involved lend the finding physiological plausibility. Microstate C is commonly associated with the salience network, which monitors internally generated and emotionally significant stimuli, while microstate D has been linked to the frontoparietal attention network, and microstate A to phonological and auditory processing. Depression has long been theorized, in cognitive models of the disorder, to involve excessive engagement of self-referential and ruminative networks at the expense of attentional control. Abnormal sequencing through salience- and attention-related microstates fits that picture: the brain’s traffic controller appears to route activity along atypical paths in MDD, and those atypical paths are stable enough across individuals and datasets to serve as a measurable signature.

The implications extend in two directions. Clinically, an EEG-based recognition model of this kind points toward objective, low-cost screening tools. Electroencephalography is non-invasive, relatively inexpensive, and already widely deployed, so a pipeline that converts a few minutes of resting-state recording into a depression risk estimate could complement, though certainly not replace, careful clinical assessment. The attention mechanism in the multiple instance learning model also offers a degree of interpretability that many deep learning approaches lack, because it highlights which mined transition rules drove a given classification, potentially giving clinicians and researchers a window into why the model reached its decision. Methodologically, the study demonstrates that sequential pattern mining, a tool rarely applied to brain dynamics, can extract discriminative structure that neither static features nor simple pairwise transition matrices capture.

Cautions remain before such a model could reach the clinic. The self-collected dataset and MODMA differ in size and recording conditions, and although performance was strong on both, larger and more diverse cohorts will be needed to confirm that the identified rules generalize across ages, medications, comorbidities, and recording hardware. Depression also overlaps biologically with other psychiatric conditions, so future work must test whether microstate transition signatures distinguish MDD from anxiety, bipolar disorder, and related diagnoses. Still, the convergence of evidence is compelling: two independent datasets, statistically corrected findings, and consistent rules pointing to the same disrupted sequences. The study, published as an original research article with the DOI 10.1007/s11517-026-03662-5, adds to a rapidly growing literature suggesting that the brain’s fastest rhythms, organized into transient states lasting barely a tenth of a second, hold clinically meaningful information about one of humanity’s most common and disabling disorders. If subsequent studies replicate and extend these results, the humble EEG cap may become a far more powerful instrument for detecting depression than anyone imagined when microstates were first described more than four decades ago.

Subject of Research: EEG microstate transition dynamics and machine learning-based recognition of major depressive disorder

Article Title: Analysis of EEG microstate transition patterns and multi-instance learning-based recognition model for depression

Article References: Analysis of EEG microstate transition patterns and multi-instance learning-based recognition model for depression. (n.d.). https://doi.org/10.1007/s11517-026-03662-5

Image Credits: AI Generated

DOI: 10.1007/s11517-026-03662-5

Keywords: EEG, microstates, depression, major depressive disorder, machine learning, multiple instance learning, RuleGrowth, pattern mining, brain dynamics, MODMA dataset, attention mechanism, biomarkers