Human Sensorimotor Resting State β Events and Aperiodic Activity Exhibit Good Test-Retest Reliability

Background

Neurological diseases are some of the most impactful diseases on daily human life, particularly those affecting sensory and motor functions, such as Parkinson’s disease. Early diagnosis of these diseases is often exceedingly difficult due to the lack of early and evident brain structure changes. Furthermore, the trajectory of disease progression and rehabilitation outcomes can be highly unpredictable. Thus, there is an urgent need for a stable and reliable clinical functional biomarker to improve the diagnosis and treatment of such conditions.

Magnetoencephalography (MEG) and electroencephalography (EEG) are non-invasive electrophysiological recording methods that can capture neural activity in cortical areas. These methods have shown substantial potential in exploring functional and structural changes in the nervous system, especially within the sensorimotor system. Recent studies have found that the β rhythm (14-30 Hz) in the sensorimotor cortex is closely related to various tasks and disease states. For instance, high-amplitude β events can predict behavioral performance and are commonly altered in sensorimotor nervous system diseases.

Source Information

This study was conducted by six researchers: Amande M. Pauls, Pietari Nurmi, Heidi Ala-Salomäki, Hanna Renvall, Jan Kujala, and Mia Liljeström. The authors are primarily from institutions such as Helsinki University Hospital, Helsinki University Medical Imaging Center, and Aalto University in Finland. The paper was published in the journal “Clinical Neurophysiology” with the title “Human sensorimotor resting state beta events and aperiodic activity show good test–retest reliability” on March 20, 2024.

Detailed Research Procedures

Subjects and Data Collection

The study investigated 50 healthy adults (age range 21-70 years), all of whom were free of existing neurological diseases and learning or language disorders. The research adhered to the Helsinki ethical guidelines and was approved by the Aalto University ethics committee. Subjects underwent two MEG recordings in a resting state, each lasting 5 minutes, with an interval of one to two weeks between sessions.

Data Acquisition and Processing

Recordings were conducted in a magnetically shielded room using a 306-channel vector-view neuromagnetometer (Megin Oy, Helsinki, Finland). Data were recorded at a sampling rate of 1 kHz using a bandpass filter of 0.03-330 Hz. Head position monitoring and behavioral control assessments ensured the subjects’ wakefulness during the recordings.

MEG Signal Processing and Parameter Extraction

MEG data were preprocessed using the time-domain signal space separation method (tSSS) to remove external noise and individual head movement compensation was performed using MaxFilter software. Data processing was conducted using the MNE-Python version 1.3 software. Data were bandpass filtered at 2-48 Hz, and power spectral density (PSD) calculations were used to extract periodic and aperiodic components.

Signal Stability

The intraclass correlation coefficient (ICC) was used to evaluate the test-retest reliability of different parameters. Signals included both periodic β events and aperiodic 1/f activity components. Signal analysis used Morlet wavelet transforms and peak channel and β frequency selection (automatic, manual, and a combination of both), yielding strongly correlated results.

Parameter Extraction and Event Detection

Across the two measurement intervals, signals’ 1/f components and β events exhibited good test-retest stability. Periodic components (ICC 0.77-0.88) and β event amplitudes (ICC 0.74-0.82) were highly stable, whereas the duration of β events showed higher variability (ICC 0.55-0.7). A 2-3 minute recording was sufficient to achieve stable results, with an automation success rate of 86% for signal analysis.

Main Findings

The study found that β events and aperiodic components in the sensorimotor cortex exhibited significant test-retest stability, especially prominent in signals from the left hemisphere. Different parameters had a minimal impact on stability, but specific filter bandwidths and amplitude thresholds did influence the results (e.g., 70-80% percentile threshold performed best). A 2-minute recording duration was adequate to obtain stable test results.

Conclusion and Significance

This study demonstrated that MEG could capture stable sensorimotor β events and aperiodic activity features in individuals at rest, which can serve as potential clinical biomarkers for neurological diseases. This stable “resting sensorimotor phenotype” holds promise for use in early diagnosis and treatment monitoring in clinical settings.

Study Highlights

1. Stability Verification: ICC confirmed the high stability of sensorimotor system β events and 1/f components in a resting state.
2. Short Recording Duration Effectiveness: Recordings of only 2-3 minutes were sufficient to achieve stable results, offering convenience for clinical applications.
3. Automation of Analysis: The proposed automated analysis method succeeded in most cases, reducing manual intervention and improving analysis efficiency.

Future Directions

Future research should further validate these parameters in clinical populations and investigate measurement consistency among different patient groups. Additionally, exploring the relationship with cardiac and movement artifacts and more detailed biological significance will help in fully understanding the potential of these characteristic biomarkers.

Conclusion

Through MEG recordings of 50 healthy subjects, this study demonstrated test-retest reliability of sensorimotor system β events and aperiodic activity in a resting state. These findings support the feasibility of the resting sensorimotor β phenotype as a potential biomarker for neurological diseases, offering a new tool for early diagnosis and treatment monitoring in clinical practice.