Neuroimaging methods group (NIMEG)

Although functional brain imaging can uncover activity patterns encompassing multiple brain regions, the interplay of these regions is usually not directly addressed. Yet, neural processing supporting cognition may dynamically recruit brain functions implemented at distinct cortical regions. These short-lived networks are formed by dynamic functional connections between the participating regions. Electro- and magnetoencephalography (EEG/MEG) measure electric brain activity with high temporal resolution. However, neither method readily provides us with a network structure; they merely show the aggregated activity of all contributing regions. The challenge is to decompose the recorded MEG/EEG data into a sparse and dynamic set of brain signal sources. The goal of this project is to tackle this challenge in a new way, surpassing current functional connectivity estimation methods, and to enable real-time tracking of these networks to allow their use in neurofeedback experiments.

In the context of this project, we also develop advanced modeling of invasive intracranial EEG measurements.

Recent preprints and publications:

1. Tronarp, F., Subramaniyam, N. P., Särkkä, S., & Parkkonen, L. (2018). Tracking of dynamic functional connectivity from MEG data with Kalman filtering. In 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (pp. 1003-1006). IEEE.
2. Subramaniyam, N. P., Tronarp, F., Särkkä, S., & Parkkonen, L. (2017). Expectation–maximization algorithm with a nonlinear Kalman smoother for MEG/EEG connectivity estimation. In EMBEC & NBC 2017 (pp. 763-766). Springer, Singapore.

Collaborators

• U. Cambridge
• Helsinki Univ. Central Hospital

We study several aspects of human cognition using MEG and EEG combined with machine learning. Many of our approaches build on brain-signal features specifically reflecting attentive and conscious processing. We also develop brain–computer interfaces both for “closed-loop” neuroscientific experimentation as well as for future clinical applications.

Recent preprints and publications:

1. Zhigalov, A., Heinilä, E., Parviainen, T., Parkkonen, L., & Hyvärinen, A. (2019). Decoding attentional states for neurofeedback: Mindfulness vs. wandering thoughts. NeuroImage, 185, 565-574.

2. Zubarev, I., & Parkkonen, L. (2018). Evidence for a general performance‐monitoring system in the human brain. Human brain mapping, 39(11), 4322-4333.

3. Halme, H. L., & Parkkonen, L. (2018). Across-subject offline decoding of motor imagery from MEG and EEG. Scientific Reports, 8(10087).

4. Zubarev, I., Zetter, R., Halme, H. L., & Parkkonen, L. (2018). Robust and highly adaptable brain-computer interface with convolutional net architecture based on a generative model of neuromagnetic measurements. arXiv preprint arXiv:1805.10981.

Collaborators:

• CEA
• INSERM

Functional brain imaging methods hold promise for producing valuable information for the diagnostics of many brain disorders; however, the application of these methods is hampered by the lack of a normative database. To this end, we are aggregating a large number of magnetoencephalographic (MEG) and functional magnetic resonance imaging (fMRI) datasets into such a database for the application of machine-learning methods to derive biomarkers that would be indicative of brain disease states.

Recent preprints and publications:

1. Maestú, F., Peña, J. M., Garcés, P., González, S., Bajo, R., Bagic, A., ... & Nakamura, A. (2015). A multicenter study of the early detection of synaptic dysfunction in Mild Cognitive Impairment using Magnetoencephalography-derived functional connectivity. NeuroImage: Clinical, 9, 103-109.

Collaborators:

• Helsinki Univ. Central Hospital
• U. Oulu
• Turku Univ. Central Hospital
• Kuopio Univ. Hospital
• CSC
• Combinostics Oy

We exploit recent advances in a novel magnetic sensor technology—optical magnetometry—to construct a new kind of MEG system that allows capturing cerebral magnetic fields within millimetres from the scalp. Our simulations show that this proximity leads up to a 5-fold increase in the signal amplitude and an order-of-magnitude improvement of spatial resolution compared to conventional SQUID-based MEG. Therefore, a high-resolution MEG (HRMEG) system based on optical magnetometers should enable non-invasive recordings of cortical activity at unprecedented sensitivity and detail level, which we capitalize on by characterizing cortical responses, particularly gamma oscillations, during complex cognitive tasks.

Read more and check the publications of this project here.

Collaborators:

• University of Colorado
• MEGIN Oy
• QuSpin Inc.
• VTT
• CSEM
• ICFO

Most neuroimaging studies of human social cognition have focused on brain activity of single subjects. “Two-person neuroimaging” refers to simultaneous recordings of brain signals from two subjects involved in social interaction. We have developed a set-up that connects two MEG systems in different laboratories allowing the subjects to interact and we have applied this set-up to study brain functions subserving social interaction.

Recent preprints and publications:

1. Mandel, A., Bourguignon, M., Parkkonen, L., & Hari, R. (2016). Sensorimotor activation related to speaker vs. listener role during natural conversation. Neuroscience letters, 614, 99-104.
2. Zhou, G., Bourguignon, M., Parkkonen, L., & Hari, R. (2016). Neural signatures of hand kinematics in leaders vs. followers: A dual-MEG study. Neuroimage, 125, 731-738.

3. Hari, R., Henriksson, L., Malinen, S., & Parkkonen, L. (2015). Centrality of social interaction in human brain function. Neuron, 88(1), 181-193.

4. Zhdanov, A., Nurminen, J., Baess, P., Hirvenkari, L., Jousmäki, V., Mäkelä, J. P., ... & Parkkonen, L. (2015). An internet-based real-time audiovisual link for dual MEG recordings. PLoS One, 10(6), e0128485.

Collaborators:

• Aarhus U.
• Helsinki Univ. Central Hospital
• U. Libre de Bruxelles / Erasmus hospital

We are contributing to open-source MEG/EEG analysis software packages.

Recent preprints and publications:

1. Gramfort, A., Luessi, M., Larson, E., Engemann, D. A., Strohmeier, D., Brodbeck, C., ... & Hämäläinen, M. S. (2014). MNE software for processing MEG and EEG data. Neuroimage, 86, 446-460.
2. Gramfort, A., Luessi, M., Larson, E., Engemann, D. A., Strohmeier, D., Brodbeck, C., ... & Hämäläinen, M. (2013). MEG and EEG data analysis with MNE-Python. Frontiers in neuroscience, 7, 267.
3. Sudre, G., Parkkonen, L., Bock, E., Baillet, S., Wang, W., & Weber, D. J. (2011). rtMEG: a real-time software interface for magnetoencephalography. Computational intelligence and neuroscience, 2011, 11.