||Active Shape Model (ASM) is a widely used statistical model-based method for recognizing and delineating structures in images. It searches images with a flexible and compact model by using prior knowledge about shape statistics derived from a training set of shapes. However, it is limited by accuracy because the segmentation results are parametric descriptions of the identified shape, and they do not often match the perceptually identified boundary in images optimally. These inaccuracies will pose problems, especially for subsequent analysis in medical imagery. On the other hand, purely image-based optimal boundary detection segmentation methods, such as live wire (LW), perform well in delineation of boundaries of objects, although they require boundary recognition help from the user. In this presentation, two strategies are proposed – live wire active shape models (LWASM) and oriented active shape models (OASM) – to actively exploit the synergy between model-based and image-based methods during segmentation. In these strategies, ASM helps LW in object recognition, and LW helps ASM in accurate delineation, and by the boundary orientedness property, in simplifying the model and in recognition itself. Evaluation based on medical image data from a variety of applications indicates that OASM yields better accuracy than LWASM, which in turn performs significantly better than ASM, although OASM is computationally more demanding than both.