三维多模态融合图像诊治颞下颌关节紊乱病的关键技术研究

The 3D multi-modality image fusion provides more medical information than any imaging modality alone. Temporomandibular joint (TMJ) is the most sophisticated joint in the body and any currently often used 3D imaging modality such as cone beam computed tomography (CBCT), computed tomography (CT) and magnetic resonance imaging (MRI) alone cannot disclose the sophisticated relationships of the normal and abnormal tissues. The purpose of the present project is to establish an image fusion platform especially for the diagnosis and treatment planning of temoromandibular joint disorders (TMD). To achieve this goal, two key factors that have a significant impact on the registration and fusion of two modality images should be developed. One is to precisely and accurately register two 3D images without a common imaging sequence. Another factor is to display minor changes of bony defects in TMJ. The results from the two factors study will lead to an establishment of an imaging fusion platform especially for TMD. The platform will be used for 1) ruling out the possible anatomical factors that give rise to TMD, the most common temporomandibular diseases, 2) amending the currently used diagnostic criteria for TMD and, 3) improving the treatment effect for patients with TMD. The results of the project are expected to demonstrate that the fusion image platform is an effective imaging platform for the early and differential diagnosis of temporamandibular joint disorders and is useful for planning individual treatment and subsequently getting an improvement on healing effect. The significance of the project is not only to explore the possible application of fusion images on the diagnosis and treatment planning of temporomandibular disorders, but also to establish a new route for diagnosis and treatment of other disease occurring in oral and maxillofacial regions.

三维多模态医学影像融合图像可提供的诊疗信息大于各单一模态所能提供信息之和。如何将不同轴扫描下获取的口腔颌面锥形束CT与磁共振图像进行精确配准、融合，如何有效而直观地显示治疗前后颞下颌关节区域正常与病变组织结构变化以及融合后的图像可否提高颞下颌关节紊乱病的诊断效能等是本课题的主要研究目的。研究内容主要是利用同一受检者的口腔锥束CT与MR图像实现面向颞下颌关节紊乱病的模板融合系统，空间融合系统以及时间融合系统的建立，并研究其在颞下颌关节紊乱病中的诊治效能。预期结果：新建立的融合系统可自动完成不同轴扫描下获取的口腔颌面锥形束CT与磁共振图像的精确配准、融合并能够提高颞下颌关节紊乱病的诊断效能。研究意义：为探索颞下颌关节紊乱病的发病因素、准确诊断、个体化治疗和进一步研发相关手术导航设备、颞下颌关节组织工程学重建修复的建模等具有重要的临床研究价值和方法创新，也为口腔医学其他疾病的诊治提供了全新途径。

研究背景：颞下颌关节是人体中解剖结构最为复杂和精细的关节之一，将上颌骨与下颌骨有机联结在一起，在生理功能上不仅完成人类赖以生存的咀嚼和吞咽活动，而且还参与言语和表情等功能。故其一旦罹患疾病，且病程迁延、反复发作、经久不愈者，将严重影响患者日常生活。如何有效、客观地对颞下颌关节疾病进行准确诊断，进而采取有效治疗措施，快速解除患者痛苦，是口腔医学临床一直追求的目标。口腔颌面锥形束CT、磁共振成像对颞下颌关节的显示各有优点，如能将不同模态的图像融合，融合后图像所获得的信息可能会大于单一模态图像，可用于探索颞下颌关节病的诊断及治疗方法。.研究内容：（1）研究健康人群中颞下颌关节髁突形态学指标，为模板建立提供形态学基础；（2）研究建立面向颞下颌关节紊乱病诊治的三维多模态医学影像融合平台，建立空间融合模块和时间融合模块；(3)研究三维多模态医学影像融合平台在诊治颞下颌关节紊乱病中的诊断效能。.研究结果：（1）已完成1010侧（505例）颞下颌关节无症状人群髁突形态学指标的测量与分类，为模板建立提供形态学基础。（2）借助Amira软件建立了一套完整的专用于颞下颌关节CBCT和MRI图像的单模态和多模态医学图像精确配准融合方法。（3）自行研制开发了针对于CBCT图像和CBCT与MRI图像的自动配准融合技术，使颞下颌关节图像的配准融合时间达到秒级水平。（4）通过建立的时间融合模块，第一次实现了三维立体图像直观清晰显示髁突术前术后位置、角度变化并进行实时测量，以及定性评价分析；通过建立的空间融合模块，对颞下颌关节疾病的诊治更加清晰、直观和准确，在观察者一致性和诊断准确性方面显著高于单一的CBCT或MRI图像。.研究意义：为探索颞下颌关节疾病的发病因素、准确诊断、个体化治疗和进一步研发相关手术导航设备、颞下颌关节组织工程学重建修复的建模等方面具有重要的临床研究价值和方法创新，也为口腔医学其他疾病的诊治提供了全新途径。