Motion-robust brain MRI for infants

Project Abstract/Summary The objective of this SBIR/STTR Fast Track submission is to provide a software solution to reduce the effects of head motion during brain MRIs of infants. Because structural MRI scanning produces high-resolution images and does not expose patients to radiation, it has become an immensely valuable diagnostic tool, particularly for imaging the brain in infant samples. Approximately $20 billion worth of brain MRIs are performed each year in the US, including 500,000 infant scans. The greatest problem for MRI is its susceptibility to patient motion. About 20% of clinical brain MRIs are distorted by motion – wasting ~$4 billion/year. Currently, MRI motion is dealt with by repeating scans or anesthesia. Both approaches diminish throughput, while increasing costs, patient burden, and health risks. In infants, anesthesia causes brain injury, leading to an FDA blackbox warning against it for children under 3 years old. Thus, the standard of brain MRI care in infants is swaddling them and hoping for the best. There are no infant-compatible MRI motion monitoring systems, despite the great need. Camera-based systems encounter critical limitations when used with infants – only FIRMMandapos;s software-only approach can work easily and reliably. Two obstacles must be overcome with support from this SBIR/STTR award to develop FIRMM-infant for research and clinical use. 1) Infant-compatible versions of motion-sensing/correcting sequences must be generated (T1w, T2w). 2) FIRMM-infant usability and its ability to improve MRI quality must be demonstrated. The proposed research focuses on delivering proof-of-concept for FIRMM-infant (Phase I) and building and validating a clinical-ready version of FIRMM-infant across vendors (Phase II). The FIRMM-infant device provides scan operators and physicians with real time motion information, and in combination with vNav ready motion sensing sequences will MR scans safer, faster, and less expensive for infant imaging.Project Narrative Magnetic resonance imaging (MRI) has unrivaled clinical utility, is non-invasive, and provides extremely high spatial resolution; however, MRIs have one severe limitation: patient motion during an MRI scan diminishes the quality of the resulting images. Importantly, there are no infant-compatible MRI motion monitoring systems, despite the great need. The goal of this application is to develop an accurate and non-invasive software-based solution for monitoring infant motion during brain MRIs.