Routine magnetoencephalography in memory clinic patients: A machine learning approach

Gouw, Alida A.; Hillebrand, Arjan; Schoonhoven, Deborah N; Demuru, Matteo; Ris, Peterjan; Scheltens, Philip; Stam, Cornelis J.

doi:10.1002/dad2.12227

Cited by 11 publications

(17 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to robustness and good reproducibility, clinically useful biomarkers need sensitivity and specificity of at least 80% or higher ( Babiloni et al, 2021 ; Colom-Cadena et al, 2020 ). Thus far, mainly conventional, single-channel power spectral measures have proven their diagnostic and predictive value in the predementia stages of AD ( Gouw et al, 2017 ; Gouw et al, 2021 ; Horvath et al, 2018 ; Hughes et al, 2019 ). While it is still entirely conceivable that the richness of brain network dynamics can help to detect or distinguish early aberrant neuronal behavior, the question arises whether progress can be made with a different type of approach, that somehow combines the best of both worlds.…”

Section: Introductionmentioning

confidence: 99%

Network-level permutation entropy of resting-state MEG recordings: A novel biomarker for early-stage Alzheimer’s disease?

Scheijbeler

Nifterick

Stam

et al. 2022

Network Neuroscience

Self Cite

View full text Add to dashboard Cite

Objective. Increasing evidence suggests that measures of signal variability and complexity could present promising biomarkers for Alzheimer’s disease (AD). Earlier studies have however been limited to the characterization of local activity. Here, we investigate whether a network version of permutation entropy could serve as a novel biomarker for early-stage AD. Methods. Resting-state source-space magnetoencephalography was recorded in 18 subjects with subjective cognitive decline (‘SCD’) and 18 subjects with mild cognitive impairment (‘MCI’). Local activity was characterized by permutation entropy (PE). Network interactions were studied using the inverted Joint Permutation Entropy (JPEinv), corrected for volume conduction. Results. The JPEinv showed a reduction of nonlinear connectivity in MCI subjects in the theta and alpha band. Local PE showed increased theta-band entropy. Between-group differences were widespread across brain regions. ROC analysis of classification of MCI versus SCD subjects revealed that a linear regression model trained on JPEinv features (78.4% [62.5–93.3%]) slightly outperformed PE (76.9% [60.3–93.4%]) and relative theta power based models (76.9% [60.4–93.3%]). Conclusion. Classification performance of theta JPEinv was at least as good as the relative theta power benchmark. The JPEinv is therefore a potential biomarker for early-stage AD, and should be explored in larger studies.

show abstract

Section: Introductionmentioning

confidence: 99%

Network-level permutation entropy of resting-state MEG recordings: A novel biomarker for early-stage Alzheimer’s disease?

Scheijbeler

Nifterick

Stam

et al. 2022

Network Neuroscience

Self Cite

View full text Add to dashboard Cite

show abstract

“…7 A). The solid performance of the theta band in our classification was not surprising, given that an increased power in the theta band is a well-documented early feature of AD [ 36, 49 ]. For the right hippocampus, good AUC values (> 0.8) were also obtained for other frequency bands, such as the absolute power in the lower alpha band ( Fig.…”

Section: Discussionmentioning

confidence: 87%

“…However, it is likely that the technical advantages of MEG are relevant for the detection of the early-stage changes in AD. Source-space MEG is relatively well-suited for capturing abnormal hippocampal oscillatory activity compared to EEG [ 36 ]. First, its higher spatiotemporal resolution enables more reliable source reconstruction, and description of deeper structures such as the hippocampus.…”

Section: Discussionmentioning

confidence: 99%

“…All were recruited from the Amsterdam Dementia Cohort (ADC) of the Alzheimer Center of the VU University Medical Center in the period of spring 2015–2018. All subjects were assessed according to a standard clinical protocol, which involved history taking, physical and neurological examination, an interview with a spouse or close family member, blood tests, 3T MRI of the brain according to a standard protocol, routine MEG [ 36 ], and a thorough neuropsychological assessment battery [ 37 ]. The diagnosis was made in a consensus meeting in which all the available clinical data were considered by a multidisciplinary team.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Oscillatory Activity of the Hippocampus in Prodromal Alzheimer’s Disease: A Source-Space Magnetoencephalography Study

Luppi

Schoonhoven

Nifterick

et al. 2022

JAD

Self Cite

View full text Add to dashboard Cite

Background: In Alzheimer’s disease (AD), oscillatory activity of the human brain slows down. However, oscillatory slowing varies between individuals, particularly in prodromal AD. Cortical oscillatory changes have shown suboptimal accuracy as diagnostic markers. We speculated that focusing on the hippocampus might prove more successful, particularly using magnetoencephalography (MEG) for capturing subcortical oscillatory activity. Objective: We explored MEG-based detection of hippocampal oscillatory abnormalities in prodromal AD patients. Methods: We acquired resting-state MEG data of 18 AD dementia patients, 18 amyloid-β-positive amnestic mild cognitive impairment (MCI, prodromal AD) patients, and 18 amyloid-β-negative persons with subjective cognitive decline (SCD). Oscillatory activity in 78 cortical regions and both hippocampi was reconstructed using beamforming. Between-group and hippocampal-cortical differences in spectral power were assessed. Classification accuracy was explored using ROC curves. Results: The MCI group showed intermediate power values between SCD and AD, except for the alpha range, where it was higher than both (p < 0.05 and p < 0.001). The largest differences between MCI and SCD were in the theta band, with higher power in MCI (p < 0.01). The hippocampi showed several unique group differences, such as higher power in the higher alpha band in MCI compared to SCD (p < 0.05). Classification accuracy (MCI versus SCD) was best for absolute theta band power in the right hippocampus (AUC = 0.87). Conclusion: In this MEG study, we detected oscillatory abnormalities of the hippocampi in prodromal AD patients. Moreover, hippocampus-based classification performed better than cortex-based classification. We conclude that a focus on hippocampal MEG may improve early detection of AD-related neuronal dysfunction.

show abstract

“…Especially for AD, numerous studies have reported spectral slowing (Fernández et al 2006; Wiesman et al 2022), increased theta activity and decreased alpha and beta activity (Huang et al 2000; Berendse et al 2000; Fernández et al 2002; Osipova et al 2005; Fernández et al 2006; Dauwels et al 2011; Haan et al 2008; Jelic et al 2000; Coben et al 1985; Jafari et al 2020; Nakamura et al 2018; Wiesman et al 2022). These spectral changes have been linked to disease progression (Jelic et al 2000; Coben et al 1985; Huang et al 2000; Babiloni et al 2009; Nakamura et al 2018) and were successfully used to differentiate AD patients from healthy controls (Gouw et al 2021).…”

Section: Discussionmentioning

confidence: 99%

Slowing of Frontocentral Beta Oscillations in Atypical Parkinsonism

Krösche

Kannenberg

Butz

et al. 2022

Preprint

View full text Add to dashboard Cite

Diagnosis of atypical parkinsonian syndromes mostly relies on clinical presentation as well as structural and molecular brain imaging. It has not been investigated whether cortical oscillatory activity exhibits features distinguishing these syndromes. Therefore, we measured resting-state magnetoencephalography in 13 patients with corticobasal syndrome, 10 patients with progressive supranuclear palsy, 23 patients with idiopathic Parkinson’s disease and 23 healthy controls. We compared spectral power as well as amplitude and frequency of power peaks between the groups.Atypical Parkinsonism was associated with spectral slowing, distinguishing both corticobasal syndrome and progressive supranuclear palsy from age-matched healthy controls and Parkinson’s disease. Patients with atypical Parkinsonism showed a shift of beta peaks (13-30 Hz) towards lower frequencies in frontal and central areas bilaterally. A concomitant increase in theta/alpha power relative to controls was observed in both atypical parkinsonian syndromes and in Parkinson’s disease.Our results demonstrate that slowing of frontocentral beta oscillations is characteristic of atypical Parkinsonism. Spectral slowing with a different topography has previously been observed in other neurodegenerative disorders, such as Alzheimer’s disease, suggesting that spectral slowing might be an electrophysiological marker of cortical neurodegeneration. As such, it might support differential diagnosis of parkinsonian syndromes in the future.HighlightsSlowing of beta oscillations distinguishes atypical parkinsonian syndromes from healthy controls and idiopathic Parkinson’s diseaseSpectral slowing is most pronounced in frontocentral areasBeta oscillations are shifted towards lower frequencies independent of their amplitude

show abstract

Routine magnetoencephalography in memory clinic patients: A machine learning approach

Cited by 11 publications

References 52 publications

Network-level permutation entropy of resting-state MEG recordings: A novel biomarker for early-stage Alzheimer’s disease?

Network-level permutation entropy of resting-state MEG recordings: A novel biomarker for early-stage Alzheimer’s disease?

Oscillatory Activity of the Hippocampus in Prodromal Alzheimer’s Disease: A Source-Space Magnetoencephalography Study

Slowing of Frontocentral Beta Oscillations in Atypical Parkinsonism

Contact Info

Product

Resources

About