McColl R W, Fleckenstein J L, Bowers J, Theriault G, Peshock R M
Department of Radiology, University of Texas Southwestern Medical Center, Dallas 75235.
Comput Med Imaging Graph. 1992 Nov-Dec;16(6):363-71. doi: 10.1016/0895-6111(92)90054-d.
Three-dimensional (3D) reconstructions of medical images can provide useful information to the radiologist, enabling delineation and spatial correlation of anatomic structures in one image rather than consecutive two-dimensional (2D) images. Magnetic resonance imaging (MRI) is known to be a powerful technique for evaluation of skeletal muscle pathology and physiology. We investigated the role of 3D reconstruction from MRI images of pathological and healthy muscle, using volume rendering. The ability to easily delineate and recognize normal and injured muscle in the 3D images were dependent upon the original contrast between normal and injured muscle, spatial resolution, and anatomic complexity in the original slices, and also on imaging parameters such as volume averaging.
医学图像的三维（3D）重建可为放射科医生提供有用信息，能够在一幅图像而非连续的二维（2D）图像中描绘解剖结构并进行空间关联。磁共振成像（MRI）是评估骨骼肌病理和生理状况的一项强大技术。我们利用容积再现技术研究了从病理和健康肌肉的MRI图像进行3D重建的作用。在3D图像中轻松描绘和识别正常及受损肌肉的能力取决于正常与受损肌肉之间的原始对比度、空间分辨率、原始切片中的解剖复杂性，以及诸如容积平均等成像参数。
