AUTOMATED IMAGE REGISTRATION FOR CLINICAL NEUROIMAGING

点击翻译按钮获取中文摘要

The basic hypothesis of our proposal is that image registration can be routinely applied in the clinical domain to improve diagnosis and management of specific neurological disorders. The long term goal of this proposal is the development of a completely automated system for registration of scans from computed axial tomography (CT). Recently completed engineering studies suggest that all registration steps can be performed within 2 min. The specific aims for Phase I are: 1) To create an engineering prototype system that can be used in a laboratory environment to confirm the validity of the engineering study. 2) To evaluate the speed, convenience and accuracy of the prototype on three specific medical indications for CT scanning (with and without contrast media.) The engineering prototype includes: a) software to receive and decode CT data, b)study-specific pre- and post-processing modules, c) a registration engine, and d) visualization tools. Phase I research will provide explicit demonstration of principle and physical testing in a hospital laboratory. This information will be used in Phase II to design a commercial strength product with testing in the emergency room and clinical environments leading toward compliance with FDA guidelines. PROPOSED COMMERCIAL APPLICATIONS: Despite the potential for improving diagnosis and management of neurological patients, image registration cannot be applied widely because of the cost and time associated with manual processing. Automation will lower the cost. Application to diagnosis and management of stroke disease and brain tumors are key application of this technology.