HARDWARE AND INSTRUMENTATION - Full Paper Hybrid Ultra-Low-Field MRI and Magnetoencephalography System Based on a Commercial Whole-Head Neuromagnetometer Panu T. Vesanen, 1 * Jaakko O. Nieminen, 1 Koos C. J. Zevenhoven, 1 Juhani Dabek, 1 Lauri T. Parkkonen, 1,2 Andrey V. Zhdanov, 1,3 Juho Luomahaara, 4,5 Juha Hassel, 4 Jari Penttil€ a, 5 Juha Simola, 2 Antti I. Ahonen, 2 Jyrki P. M€ akel€ a, 3 and Risto J. Ilmoniemi 1 Ultra-low-field MRI uses microtesla fields for signal encoding and sensitive superconducting quantum interference devices for signal detection. Similarly, modern magnetoencephalogra- phy (MEG) systems use arrays comprising hundreds of super- conducting quantum interference device channels to measure the magnetic field generated by neuronal activity. In this article, hybrid MEG-MRI instrumentation based on a commercial whole- head MEG device is described. The combination of ultra-low-field MRI and MEG in a single device is expected to significantly reduce coregistration errors between the two modalities, to sim- plify MEG analysis, and to improve MEG localization accuracy. The sensor solutions, MRI coils (including a superconducting polarizing coil), an optimized pulse sequence, and a reconstruc- tion method suitable for hybrid MEG-MRI measurements are described. The performance of the device is demonstrated by presenting ultra-low-field-MR images and MEG recordings that are compared with data obtained with a 3T scanner and a com- mercial MEG device. Magn Reson Med 69:1795–1804, 2013. V C 2012 Wiley Periodicals, Inc. Key words: ultra-low-field MRI; magnetoencephalography; MEG-MRI; superconducting quantum interference device INTRODUCTION Functional neuroimaging technologies, such as fMRI, positron emission tomography, electroencephalography, and magnetoencephalography (MEG), are vitally impor- tant tools in modern neuroscience. MEG (1) is based on a direct measurement of the magnetic field generated by neuronal activity, as opposed to fMRI (2,3), which meas- ures hemodynamic changes in the working human brain. However, the spatial resolution of MEG is limited, and the method offers little anatomical information about the brain, necessitating its combination with structural imag- ing. In this article, we address the challenge of combined structural and neuromagnetic measurements by describ- ing a hybrid device capable of both MRI and MEG. In MEG, the weak magnetic fields generated by neuro- nal activity are measured with highly sensitive supercon- ducting quantum interference devices (SQUIDs) coupled to superconducting receiver, or pick-up, coils. A typical MEG instrument comprises several hundred such sensors in a helmet-shaped configuration. Neuronal sources underlying a given multichannel MEG measurement can be estimated by imposing certain physiological con- straints and solving an inverse problem incorporating the sensitivity patterns of the sensors. The estimated sources are typically visualized on a structural MR image of the subject, which needs to be acquired separately. The necessary coregistration of MEG and MRI is suscep- tible to measurement errors, possible movements of the brain inside the skull between the two scans (particu- larly in patients), and distortions in the MR image. Still, thanks to its excellent temporal resolution and ability to locate neuronal activity, MEG is a widely used neuro- science research tool with clinically approved applica- tions, e.g., in the presurgical evaluation of epilepsy and brain-tumor patients (4). Recently, MRI using SQUIDs has been demonstrated (5,6). This approach, called ultra-low-field (ULF) MRI, uses microtesla-range fields for signal encoding. How- ever, for sufficient spin polarization, the prepolarization technique (7) is used. With SQUID detection, the polariz- ing field strength has typically been below 150 mT (5,6,8); with induction-coil detection, polarizing field strengths 20–500 mT have been reported (9–14). As SQUID-based sensors measure the magnetic field directly, as opposed to its time derivative, the signal-to- noise ratio (SNR) of the measurement for untuned sen- sors is independent of the Larmor frequency and thus field strength after the prepolarization (15). Many con- ventional MRI techniques such as spin-echo, Fourier- imaging, and phased-array methods are also available in the ULF regime. Furthermore, ULF MRI shares benefits of other types of low-field and prepolarized MRI: 1 Department of Biomedical Engineering and Computational Science, Aalto University School of Science, Espoo, Finland. 2 Elekta Oy, Helsinki, Finland. 3 BioMag Laboratory, HUSLAB, Helsinki University Central Hospital, Helsinki, Finland. 4 VTT Technical Research Centre of Finland, Espoo, Finland. 5 Aivon Oy, Espoo, Finland. Grant sponsor: The European Community’s Seventh Framework Programme (FP7/2007–2013); Grant number: 200859; Grant sponsors: The Academy of Finland, Emil Aaltonen Foundation, Instrumentarium Science Foundation, and International Doctoral Programme in Biomedical Engineering and Medical Physics (iBioMEP). *Correspondence to: Panu T. Vesanen, M.Sc., Department of Biomedical Engineering and Computational Science, Aalto University School of Science, P.O. Box 12200, FI-00076 AALTO, Finland. E-mail: panu.vesanen@aalto.fi Received 14 February 2012; revised 8 May 2012; accepted 24 June 2012. DOI 10.1002/mrm.24413 Published online 17 July 2012 in Wiley Online Library (wileyonlinelibrary. com). Magnetic Resonance in Medicine 69:1795–1804 (2013) V C 2012 Wiley Periodicals, Inc. 1795