Personalized X-Ray 3-D Reconstruction of the Scoliotic Spine From Hybrid Statistical and Image-Based Models

摘要: This paper presents a novel 3-D reconstruction method of the scoliotic spine using prior vertebra models with image-based information taken from biplanar X-ray images. We first propose global modeling approach by exploiting curve reconstructed coronal and sagittal image in order to generate an approximate statistical model database patients based on transformation algorithm which incorporates intuitive geometrical properties. The personalized is then achieved segmentation takes into account variable appearance vertebrae (rotation, wedging) standard quality images segment isolate individual radiographic planes. More specifically, it uses regulated 2-D level set functionals identify match corresponding bone structures X-rays. An iterative optimization procedure integrating similarity measures such as deformable vertebral contours high-level anatomical primitives, morphological knowledge epipolar constraints applied globally refine landmarks each spine. was validated twenty comparing results manual approach. qualitative evaluation retro-projection confirms that proposed can achieve better consistency image's natural content. A comparison synthetic real patient data also yields good accuracy localization low-level primitives identified expert vertebra. experiments reported this demonstrate offers matching views when compared technique for evaluated cases, its precision comparable generated magnetic resonance images, thus suitable routine clinical assessment spinal deformities.