The automatic extraction of anatomical structure contours from medical images is a challenging task in也已 presence of missing or unrelated par的， occlusions caused by other strnctures， and image noise. Employing prior information about也已anatomical strnctures has been one of也已most popular ways of addressing也已se challenges. This paper presents a novel framework也at incorporates both shape and image priors into theωntour extraction process with deformable contours.立起framework handles the deformable contour evolution and the prior information integration separately by stopping由巴巴volution of the deformable model and regularly re-initializing由巴巴xpert-delineated contour with也已most similar image and shape properties. The method can be used with any deformable model without complicating也已deformable model fnnctional. An explicit training phase is not間quired for也已construction of也已 prior model. Morωver， it can be applied to any medical shape contour extraction task with simple modifications. The system is test吋on echocardiographic images and cardiac magnetic resonance imaging (MRI) slicεs for left ventricle border extraction under the level set 企amework.The visual and numerical results show the effectiveness of也已 proposed method.