This program supports regulatory guidance by developing methods to validate dynamic neurodiagnostic biomarkers achieved using functional magnetic resonance imaging (fMRI) or quantitative electroencephalography (qEEG). These biomarkers can be used clinically to diagnose neurological diseases and disorders, assist in presurgical planning, and monitor therapeutic treatment.
Functional MRI is a non-invasive neuroimaging technology that yields an indirect measure of brain activation using the endogenous contrast of oxygenated hemoglobin. One current clinical application of fMRI is its use during presurgical planning to identify eloquent cortex to spare during the resection of tumors. However, issues such as head motion, neurovascular uncoupling, and variability in subject performance create uncertainties in the understanding of fMRI activation maps that could affect its clinical use. Current projects include researching methods to reduce noise, creating digital reference objects (DROs) to model the effect of different sources of variance on fMRI activation maps, and determining quality assurance (QA) metrics to help validate fMRI as a clinical biomarker.
EEG is an inexpensive, non-invasive neurophysiological technology that records electrical activity on the scalp. Quantitative EEG (qEEG) uses modern computational power to perform power spectral analyses of EEG time-series data. There is interest in using qEEG as a clinical biomarker to assist in the diagnosis of Alzheimer’s disease, Attention deficit hyperactivity disorder (ADHD), schizophrenia, and traumatic brain injury (TBI). Current projects include researching the reproducibility of the power spectral density bands to help validate qEEG as a clinical biomarker.
(Clockwise from the left): A montage of fMRI activation maps comparing different physiological noise reduction techniques, a map of EEG electrode placement, and a plot of EEG power spectral density.
This research is important for the agency’s regulatory mission because both fMRI and qEEG have the potential to serve as clinical biomarkers in the development of validation tools in the pre-market evaluation of the effectiveness of medical device submissions.
David Soltysik, Ph.D.
Eugene Civillico, Ph.D.
Sunder Rajan, Ph.D.
National Institutes of Health (NIH)
Quantitative Imaging Biomarkers Alliance (QIBA)
Selected peer-review publications
- Soltysik et al., Improving the use of principal component analysis to reduce physiological noise and motion artifacts to increase the sensitivity of task-based fMRI. J Neurosci Methods 2015
- Soltysik et al., Head-repositioning does not reduce the reproducibility of fMRI activation in a block-design motor task, Neuroimage 2011