Non-rigid image registration (NIR) is an essential tool for morphologic comparisons in the presence of intra and inter-individual anatomic variations. Many NIR methods have been developed, but are especially difficult to evaluate since point-wise inter-image correspondence is usually unknown. We propose to develop and test a framework for comprehensive NIR method evaluation that does not require a """"""""Gold Standard"""""""" or ground truth correspondence map. The Non-rigid Image Registration Evaluation Project (NIREP) will develop software tools and provide shared image validation databases for rigorous testing of non-rigid image registration algorithms. NIREP will extend the scope of prior validation projects by developing evaluation criteria and metrics using large image populations, using richly annotated image databases, using computer simulated data, and increasing the number and types of evaluation criteria. The goal of this project is to establish, maintain, and endorse a standardized set of relevant benchmarks and metrics for performance evaluation of non-rigid image registration algorithms. Furthermore, these standards will be incorporated into an exportable computer program to automatically evaluate the registration accuracy of non-rigid image registration algorithms. The R21 Specific Aims of this project are to: (1) Construct a richly annotated human brain MR image database to evaluate and benchmark mono-modality non-rigid image registration performance. (2) Develop non-rigid image registration performance metrics and establish small deformation non-rigid registration benchmarks. (3) Develop an exportable Linux executable program to evaluate 2D and 3D non-rigid registration performance on a high-performance personal computer. The R33 Specific Aims of this project are to: (1) Expand the shareable reference database for non-rigid registration evaluation. (2) Establish large deformation registration benchmarks and a framework for adding evaluation metrics. (3) Disseminate NIREP software and annotated reference databases to the research community. (4) Test performance of major non-rigid image registration algorithms contained in ITK and SPM using the NIREP software.
Showing the most recent 10 out of 12 publications
