Sensor-level MEG combined with machine learning yields robust classification of mild traumatic brain injury patients

The Use of Magnetoencephalography in the Diagnosis and Monitoring of Mild Traumatic Brain Injuries and Post-Concussion Syndrome

Background/Objectives: The main objective of this systematic review was to explore the role of magnetoencephalography (MEG) in the diagnosis, assessment, and monitoring of mild traumatic brain injury (mTBI) and post-concussion syndrome (PCS). We aimed to evaluate the potential of some MEG biomarkers in detecting subtle brain abnormalities often missed by conventional imaging techniques. Methods: A systematic review was conducted using 25 studies that administered MEG to examine mTBI and PCS patients. The quality of the studies was assessed based on selection, comparability, and outcomes. Studies were analyzed for their methodology, evaluated parameters, and the clinical implications of using MEG for mTBI diagnosis. Results: MEG detected abnormal brain oscillations, including increased delta, theta, and gamma waves and disruptions in functional connectivity, particularly in the default mode and frontoparietal networks of patients suffering from mTBI. MEG consistently revealed abnormalities in mTBI patients even when structural imaging was normal. The use of MEG in monitoring recovery showed significant reductions in abnormal slow-wave activity corresponding to clinical improvements. Machine learning algorithms applied to MEG data demonstrated high sensitivity and specificity in distinguishing mTBI patients from healthy controls and predicting clinical outcomes. Conclusions: MEG provides a valuable diagnostic and prognostic tool for mTBI and PCS by identifying subtle neurophysiological abnormalities. The high temporal resolution and the ability to assess functional brain networks make MEG a promising complement to conventional imaging. Future research should focus on integrating MEG with other neuroimaging modalities and standardizing MEG protocols for clinical use.

Disconnection of alpha oscillations within default mode network associated with memory dysfunction in amnestic MCI

OBJECTIVE

Episodic memory dysfunction and alterations of functional connectivity (FC) in default mode network (DMN) were found in patients with amnestic mild cognitive impairment (aMCI). However, previous studies were limited in probing certain oscillations within the DMN. This study employed measures of resting-state FC across various oscillations within the DMN to comprehensively examine the FC and its association with episodic memory performance in aMCI.

METHODS

Twenty-six healthy controls (HC) and 30 patients with aMCI were recruited to perform resting-state magnetoencephalographic recordings. We compared the spectral powers and peak frequency values in each frequency band and FC within the DMN between these two groups. The associations of FC values with memory performance were also examined.

RESULTS

No significant between-group differences in spectral powers and peak frequency values were observed in the regional nodes. Patients with aMCI exhibited diminished alpha-band FC as compared to HC. Furthermore, lower alpha-band FC between the medial temporal cortex - and the posterior cingulate cortex/precuneus was correlated with poorer memory performance.

CONCLUSIONS

Aberrant DMN connectivity, particularly in the alpha frequency range, might be a neural correlate of episodic memory impairment.

SIGNIFICANCE

Our results inform the potential development of brain stimulation in managing memory impairments in aMCI.

From bench to bedside: Overview of magnetoencephalography in basic principle, signal processing, source localization and clinical applications

Magnetoencephalography (MEG) is a non-invasive technique that can precisely capture the dynamic spatiotemporal patterns of the brain by measuring the magnetic fields arising from neuronal activity along the order of milliseconds. Observations of brain dynamics have been used in cognitive neuroscience, the diagnosis of neurological diseases, and the brain-computer interface (BCI). In this study, we outline the basic principle, signal processing, and source localization of MEG, and describe its clinical applications for cognitive assessment, the diagnoses of neurological diseases and mental disorders, preoperative evaluation, and the BCI. This review not only provides an overall perspective of MEG, ranging from practical techniques to clinical applications, but also enhances the prevalent understanding of neural mechanisms. The use of MEG is expected to lead to significant breakthroughs in neuroscience.

Using normative modeling and machine learning for detecting mild traumatic brain injury from magnetoencephalography data

New biomarkers are urgently needed for many brain disorders; for example, the diagnosis of mild traumatic brain injury (mTBI) is challenging as the clinical symptoms are diverse and nonspecific. EEG and MEG studies have demonstrated several population-level indicators of mTBI that could serve as objective markers of brain injury. However, deriving clinically useful biomarkers for mTBI and other brain disorders from EEG/MEG signals is hampered by the large inter-individual variability even across healthy people. Here, we used a multivariate machine-learning approach to detect mTBI from resting-state MEG measurements. To address the heterogeneity of the condition, we employed a normative modeling approach and modeled MEG signal features of individual mTBI patients as deviations with respect to the normal variation. To this end, a normative dataset comprising 621 healthy participants was used to determine the variation in power spectra across the cortex. In addition, we constructed normative datasets based on age-matched subsets of the full normative data. To discriminate patients from healthy control subjects, we trained support-vector-machine classifiers on the quantitative deviation maps for 25 mTBI patients and 20 controls not included in the normative dataset. The best performing classifier made use of the full normative data across the entire age and frequency ranges. This classifier was able to distinguish patients from controls with an accuracy of 79%. Inspection of the trained model revealed that low-frequency activity in the theta frequency band (4-8 Hz) is a significant indicator of mTBI, consistent with earlier studies. The results demonstrate the feasibility of using normative modeling of MEG data combined with machine learning to advance diagnosis of mTBI and identify patients that would benefit from treatment and rehabilitation. The current approach could be applied to a wide range of brain disorders, thus providing a basis for deriving MEG/EEG-based biomarkers.