IMPROVED BRAIN SPECT IMAGING

Electrosonics Medical Inc, Cleveland OH

Investigators

Abstract

Nuclear imaging (SPECT and PET) is the most commonly used modality for functional brain imaging. However, a major limitation to neurological SPECT imaging is spatial resolution. Brain imaging requires optimal spatial resolution because of the intricate structures involved and the accuracy needed for quantitative mapping. Systematic errors from both patient motion and mechanical misalignment of the detectors contribute to lost spatial resolution and could produce artifacts that hinder diagnosis. Patient motion is a significant issue due to the long acquisition times (20 minutes or more), especially for pediatric patients and for patients with neurological disorders. In Phase l, we will investigate a novel method for determining detector misalignment and patient motion to improve spatial resolution and to eliminate artifacts. Our method will account for more degrees of freedom than other methods and will have higher accuracy. To demonstrate the accuracy of our approach, we will measure the uncertainties in the measured linear and angular displacements, and we will characterize the ability to recover spatial resolution using 3D reconstruction techniques. This research will lead to a standalone commercial product that can be utilized with virtually any SPECT imaging system. PROPOSED COMMERCIAL APPLICATIONS: We expect that this research will lead to a standalone commercial product to augment brain SPECT imaging by improving spatial resolution and eliminating artifacts in images. The product may be bundled with new SPECT systems and also sold separately for already-installed systems. Our vision is that through further research in advanced image processing and reconstruction, the product will evolve into a specialized quantitative neurological SPECT analysis package. This research may also find commercial acceptance as a technique for camera acceptance testing and quality assurance.