Quality of fit of stable mass-spring models

Shape models are used for many tasks in modern image analysis, e.g., segmentation, tracking, etc. Rarely, the quality of fit of adapted shape models is automatically determined to decide whether the model fitting was successful. This paper develops a principal strategy to measure the success of the model adaptation and defines a quality of fit measure for Stable Mass-Spring Models, which are especially appropriate for this purpose. The parameters of this measure can be estimated from a successfully fitted model instance. The quality of fit measure was tested with two medical segmentation tasks (left ventricles in SPECT data and lymph nodes in CT data) and could be shown to work well. Using application-specific thresholds on the quality of fit, we were able to detect automatically whether a segmentation succeeded or failed. © 2008 Pleiades Publishing, Ltd.