In this project we will install a magnetic resonance imaging (MRI) compatible head motion tracking system in an existing 3 T Prisma MRI scanner, and it will be available for use with the other MRI systems (including the 7 T and MR/PET systems). The tracker will enable passive motion monitoring for efficient instruction and rescanning of patients and volunteers, offline motion regression in research data, and retrospective image correction. The system also enables real-time motion correction during brain scans. It has been observed that head movement during brain scans affects clinical diagnosis, especially in the sickest patients, and costs hospitals approximately $115,000 per scanner per year. At the Martinos Center, hundreds of research scans are performed every month on the Prisma scanner, and the proposed motion tracking instrument promises to improve image quality and imaging success rate in many of these studies. The proposed instrument is unique in that it uses an external camera but does not require any markers. The use of a camera ensures that head position estimates are generated rapidly and independent of the imaging sequence. The need for a marker is eliminated through use of a structured light approach that enables the system to model facial geometry and track head movement directly without touching the patient. Eliminating the marker also reduces setup time and greatly improves reliability relative to other camera-based systems. The Martinos Center scanners, and especially the Prisma scanner, are used in many NIH-funded research studies, ranging from basic neuroscience to clinical research. While almost every brain imaging study can benefit from the tracking system, we highlight twelve specific NIH-funded projects run by nine PIs who are excited to incorporate the system in their imaging studies and that represent a range of applications and scanners. The highlighted studies focus on (1) spectroscopic glioma imaging, (2) imaging for automated morphometry, (3) imaging networks in Alzheimer?s patients, (4) glial activation and pain after knee replacement due to osteoarthritis, and lower back pain, (5) combined multichannel TMS and MRI, (6) microvascular imaging and modeling, (7) imaging to understand cerebrovascular contributions to aging and dementia, and (7) imaging of neonates in a study of the effects of HIV and ART exposure on the developing brain. The Executive Committee on Research (ECOR) of MGH has dedicated substantial additional funds for personnel and system maintenance, so that researchers can improve their studies using the proposed instrument.
Patient motion during magnetic resonance imaging has been estimated to cost hospitals $115,000 per scanner per year, and disproportionally affects quality of diagnosis in the sickest patients, while motion during research brain scans may bias study conclusions. Conventional external tracking systems require a marker attached to the patient?s face. The proposed TCL3.02 markerless tracking system uses structured infrared light and facial geometry to provide 30 head position estimates per second, without touching the patient. Passive tracking informs the technician whether to stop and repeat the scan, and provides regression information for researchers to analyze experimental data, while active real-time correction during acquisitions is also supported.
