Graphics and Geomatics Group, University of Jaén, Jaén, Spain.
Comput Methods Programs Biomed. 2017 Apr;141:93-104. doi: 10.1016/j.cmpb.2016.12.014. Epub 2016 Dec 31.
The preoperative planning of bone fractures using information from CT scans increases the probability of obtaining satisfactory results, since specialists are provided with additional information before surgery. The reduction of complex bone fractures requires solving a 3D puzzle in order to place each fragment into its correct position. Computer-assisted solutions may aid in this process by identifying the number of fragments and their location, by calculating the fracture zones or even by computing the correct position of each fragment. The main goal of this paper is the development of an automatic method to calculate contact zones between fragments and thus to ease the computation of bone fracture reduction.
In this paper, an automatic method to calculate the contact zone between two bone fragments is presented. In a previous step, bone fragments are segmented and labelled from CT images and a point cloud is generated for each bone fragment. The calculated contact zones enable the automatic reduction of complex fractures. To that end, an automatic method to match bone fragments in complex fractures is also presented.
The proposed method has been successfully applied in the calculation of the contact zone of 4 different bones from the ankle area. The calculated fracture zones enabled the reduction of all the tested cases using the presented matching algorithm. The performed tests show that the reduction of these fractures using the proposed methods leaded to a small overlapping between fragments.
The presented method makes the application of puzzle-solving strategies easier, since it does not obtain the entire fracture zone but the contact area between each pair of fragments. Therefore, it is not necessary to find correspondences between fracture zones and fragments may be aligned two by two. The developed algorithms have been successfully applied in different fracture cases in the ankle area. The small overlapping error obtained in the performed tests demonstrates the absence of visual overlapping in the figures.
通过 CT 扫描获取的骨折信息进行术前规划，可以提高获得满意结果的概率，因为这为手术前的专家提供了更多信息。为了将每块碎骨放置在正确的位置，需要解决一个三维拼图，以还原复杂的骨折。计算机辅助解决方案可以通过识别碎片的数量和位置、计算骨折区域甚至计算每个碎片的正确位置来辅助这一过程。本文的主要目的是开发一种自动方法来计算碎片之间的接触区域，从而简化骨折复位的计算。
本文提出了一种自动计算两块骨头之间接触区域的方法。在之前的步骤中，从 CT 图像中分割和标记骨头碎片，并为每个骨头碎片生成一个点云。计算出的接触区域可以实现复杂骨折的自动复位。为此，还提出了一种自动匹配复杂骨折中骨头碎片的方法。
该方法已成功应用于计算踝关节区域的 4 块不同骨头的接触区域。计算出的骨折区域使所有测试病例都可以使用提出的匹配算法进行复位。所进行的测试表明，使用提出的方法对这些骨折进行复位会导致碎片之间有较小的重叠。
所提出的方法使得拼图策略的应用更加容易，因为它不是获取整个骨折区域，而是获取每对碎片之间的接触区域。因此，不需要在碎片之间找到骨折区域的对应关系，可以逐个对齐。所开发的算法已成功应用于踝关节区域的不同骨折病例。在进行的测试中，得到的重叠误差较小，证明了图像中不存在明显的重叠。
