Neuro Ocillo Charts: Charting Brain Oscillations using Normative Modeling
Poster No:
979
Submission Type:
Abstract Submission
Authors:
Mohammad Zamanzadeh Nasrabadi1, Ymke Verduyn1, Augustijn de Boer2, Tomas Ros3, Thomas Wolfers4, Richard Dinga1, Marie Šafář1, Marijn van Wingerden1, Andre Marquand2, Seyed Kia1
Institutions:
1Tilburg University, Tilburg, Brabant, 2Donders Institute for Brain, Cognition and Behaviour, Nijmegen, Gelderland, 3Center for Biomedical Imaging, Geneva University Hospitals, Geneva, Switzerland, Geneva, Canton of Geneva, 4University of Tübingen, Tübingen, Baden-Württemberg
First Author:
Co-Author(s):
Tomas Ros, dr.
Center for Biomedical Imaging, Geneva University Hospitals, Geneva, Switzerland
Geneva, Canton of Geneva
Center for Biomedical Imaging, Geneva University Hospitals, Geneva, Switzerland
Geneva, Canton of Geneva
Introduction:
Normative modeling (NM) (Marquand et al., 2019) has recently gained attention for characterizing heterogeneity in neuroimaging cohorts. Here, we aim to derive lifespan growth charts for brain oscillations on large, multi-site (5 sites), and multi-device (3 manufacturers) Magnetoencephalography (MEG) data and use them to characterize the individual-level deviations of Parkinson's patients from the population norm.
Methods:
We used resting-state MEG data from 1,840 healthy participants and 160 patients with Parkinson's disease (PD), aged 6 to 88 years. Data were pooled from five datasets (Cam-CAN (Taylor et al., 2017), Boys Town National Research Hospital (Rempe et al., 2023), OMEGA (Niso et al., 2016), Human Connectome Projects (Van Essen et al., 2012) and National Institutes of Mental Health (Nugent et al., 2022)). After segmentation and averaging over time, the power spectral density (PSD) of segments was computed. Periodic components were then extracted from the PSDs using the SpecParam algorithm (Gerster et al., 2022). Then, relative frequency powers of theta, alpha, beta, and gamma bands were extracted and averaged across channels. We used hierarchical Bayesian regression (HBR) (Kia et al., 2022) with a SHASH likelihood (de Boer et al., 2024) and B-spline bases to model heteroscedastic and non-linear trajectories of brain oscillations, using sex and site as grouping effects. The NMs are estimated on the healthy training data (50% of data) and then evaluated on the test set.
We used the 50th centiles of the growth charts to derive Population-level Neuro Oscillo Charts (P-NOCs) to delineate the changes in the power of different frequency bands with aging. We further introduce Individual-level Neuro Oscillo Charts (I-NOCs), a tool to benchmark individuals' measurements against a large population's norms.
As a preliminary clinical validation and in an anomaly detection scenario, we benchmarked the models on PD patients. To this end, the predictive power of deviations, derived from NMs, was tested by calculating the Abnormal Probability Index (Kia et al., 2022) of Parkinson's patients and 220 healthy subjects from the test set.
We used the 50th centiles of the growth charts to derive Population-level Neuro Oscillo Charts (P-NOCs) to delineate the changes in the power of different frequency bands with aging. We further introduce Individual-level Neuro Oscillo Charts (I-NOCs), a tool to benchmark individuals' measurements against a large population's norms.
As a preliminary clinical validation and in an anomaly detection scenario, we benchmarked the models on PD patients. To this end, the predictive power of deviations, derived from NMs, was tested by calculating the Abnormal Probability Index (Kia et al., 2022) of Parkinson's patients and 220 healthy subjects from the test set.
Results:
Figure 1(a) presents the estimated growth charts for the relative power of four frequency bands in males and females, spanning ages 5 to 80. The wide range of variation in the derived centiles highlights substantial variability within the general population. The P-NOCs in Figure 1(b) reveal an increase in relative alpha power until around age 20, followed by a gradual decline until 80. The relative theta power follows an inverted U-shape, with a nadir at age 45, while the relative beta power displays a continuously increasing trend across the lifespan. Figure 1(c) depicts an I-NOCs for a healthy participant with normal (within the interquartile) brain oscillation across four frequency bands.
Figure 2(a) presents the result of anomaly detection on the OMEGA dataset. It illustrates the AUCs in detecting Parkinson's patients from healthy participants in the test set. The AUCs suggest that Parkinson's patients showed higher anomalies in the Theta and Beta bands. Figure 2(b) displays a scatter plot of the distribution of Parkinson's patients in Theta-Beta z-score space that highlights the heterogeneity in the patients as a spectrum from high Theta-low Beta to low Theta-high Beta deviations. The I-NOCs illustrate heterogeneous profiles of seven example patients across this spectrum.
Figure 2(a) presents the result of anomaly detection on the OMEGA dataset. It illustrates the AUCs in detecting Parkinson's patients from healthy participants in the test set. The AUCs suggest that Parkinson's patients showed higher anomalies in the Theta and Beta bands. Figure 2(b) displays a scatter plot of the distribution of Parkinson's patients in Theta-Beta z-score space that highlights the heterogeneity in the patients as a spectrum from high Theta-low Beta to low Theta-high Beta deviations. The I-NOCs illustrate heterogeneous profiles of seven example patients across this spectrum.
Conclusions:
We modeled centiles of variation of neural oscillations across the human lifespan using a large-scale and multi-site MEG dataset. We used the estimated centiles to derive individual-level Neuro Oscillo Charts that can be possibly used as an instrument for personalizing diagnosis and treatment in neuropsychiatry.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s)
Lifespan Development:
Aging
Early life, Adolescence, Aging 1
Modeling and Analysis Methods:
Bayesian Modeling
EEG/MEG Modeling and Analysis 2
Keywords:
Aging
Data analysis
Machine Learning
MEG
Modeling
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
Which processing packages did you use for your study?
Provide references using APA citation style.
Gerster, M. (2022). Separating neural oscillations from aperiodic 1/f activity: challenges and recommendations. Neuroinformatics, 20(4), 991–1012.
Kia, S. M. (2022). Closing the life-cycle of normative modeling using federated hierarchical bayesian regression. Plos One, 17(12), e0278776.
Marquand, A. F. (2019). Conceptualizing mental disorders as deviations from normative functioning. Molecular Psychiatry, 24(10), 1415–1424.
Niso, G. (2016). Omega: the open meg archive. Neuroimage, 124, 1182–1187.
Nugent, A. C. (2022). The nimh intramural healthy volunteer dataset: A comprehensive meg, mri, and behavioral resource. Scientific Data, 9(1), 518.
Rempe, M. P. (2023). Spontaneous cortical dynamics from the first years to the golden years. Proceedings of the National Academy of Sciences, 120(4), e2212776120.
Taylor, J. R. (2017). The cambridge centre for ageing and neuroscience (cam-can) data repository: Structural and functional mri, meg, and cognitive data from a cross-sectional adult lifespan sample. Neuroimage, 144, 262–269.
Van Essen, D. C. (2012). The human connectome project: a data acquisition perspective. Neuroimage, 62(4), 2222–2231.