The goal of this SBIR project is to design a low-profile, high-resistive, MR-compatible dense-array EEG (dEEG) sensor net for simultaneous dEEG/fMRI recordings in fields as high as 7 Tesla. This novel sensor net, the InkNet, will provide safe, noninvasive, and affordable dEEG/fMRI technology to both clinicians and researchers, thereby enabling routine multimodal imaging of human brain function with unprecedented spatiotemporal resolution. Application of this technology will enhance basic science of healthy brain function, as well as treatment of many neural pathologies and pre-surgical planning. The InkNet will overcome current cross-modal safety and artifact issues that have so far severely limited the effectiveness of simultaneous dEEG/fMRI by leveraging expertise in innovative polymer thick film (PTF) technology at the A. A. Martinos Center, Massachusetts General Hospital and dEEG sensor net design and technology expertise at Electrical Geodesics Inc. (EGI). In Phase I, we established feasibility with the development and testing of our first 256- channel InkNet prototype using screen-printed high-resistive PTF ink leads interfaced with EGI's patented geodesic net structure and MR-compatible EEG acquisition hardware and software. Phase II will build on the successes of our Phase I prototype while working to refine its design, enhance production manufacturability and cost efficiency, and conduct performance and safety tests.
Specific Aim 1 is to study the latest innovations in flexible and stretchable substrates and conductive inks to improve conformability and electrode contact, and reduce MR-induced ballistocardiogram artifact. Sample circuits using the best candidate materials will be printed in-house and put to rigorous testing for optimal performance in high MR fields.
Specific Aim 2 will refine the InkNet design, including an electrode pedestal with an ultra low profile of d4 mm to fit in tight MR head coils, and a novel lead layout enabled by state-of-the-art inkjet printing of leads up to 2.5 meters in length, double-sided, 5-mil trace width, and te novel, flexible materials.
Specific Aim 3 will implement the new design and materials to produce Phase II 256-channel InkNet prototypes using QA production and test procedures and new, custom-designed assembly and test fixtures to enhance speed and reliability.
Specific Aim 4 will test and validate the Phase II prototype for human safety and data integrity. Safety tests will be performed using finite difference time domain (FDTD) numerical simulations with an anatomically accurate head model, followed by actual temperature measurements using a specially developed phantom (CHEMA) and high-power TSE imaging sequences that induce RF heating. After confirming safety, MRI and EEG data integrity will be tested at 3T and 7T field strengths using clinically relevant structural scans, and fMRI/EEG resting, visual and auditory protocols. Data quality will be compared to data from EEG-only and MRI-only sessions, and against data from a commercial MR-compatible net built with traditional copper wire technology. Our test plan will be reviewed with FDA and adjusted where required to meet requirements for a 510K predicate application.
The goal of this project is to develop a system for simultaneous measurement of brain activity using two complementary methods: electroencephalography (EEG) and functional magnetic resonance imaging (fMRI). This state-of-the-art system will offer brain scientists and clinicians a safe, non-invasive tool for studying human brain function with unprecedented spatial and temporal precision. This knowledge will help us better understand healthy brain function, treat many disorders (e.g., epilepsy), and improve pre-surgical planning.
| Poulsen, Catherine; Wakeman, Daniel G; Atefi, Seyed Reza et al. (2017) Polymer thick film technology for improved simultaneous dEEG/MRI recording: Safety and MRI data quality. Magn Reson Med 77:895-903 |
| Krishnaswamy, Pavitra; Bonmassar, Giorgio; Poulsen, Catherine et al. (2016) Reference-free removal of EEG-fMRI ballistocardiogram artifacts with harmonic regression. Neuroimage 128:398-412 |
