基于多模态影像信息的搏动性耳鸣机制研究

搏动性耳鸣（PT）机制不清，是影响人群健康的重大医学难题。近年来相关机制研究陷入瓶颈，关键是缺乏针对性医学影像定量信息研究方法。基于团队先验成果，率先提出“PT发生机制是血流、骨质、传导多因素联动，脑改变机制是耳鸣脑中枢化”全新假说。将通过1、基于多模态影像技术动态评估PT及健康人颞骨区血管及骨质形态学、血流动力学及声学多维度信息，阐明上述信息在PT发生、传导中作用及关联，探究颞骨区异常血流及血管旁骨壁缺失形成机制；2、开展影像-形态-血流-声学多物理场耦合研究，采用多变量分析与机器学习方法构建PT病因-治疗-预后一体化量化评价模型，并多层次验证PT多因素联动机制；3、基于多模神经影像数据，利用模式识别的判别式分析探索脑中枢化关键功能连接通路变化特点，建立PT脑网络特征模型，揭示PT脑改变机制。最终建立PT精准、个性化诊疗模拟系统，为PT风险预测、治疗策略及疗效预判提供科学依据及评价手段。

The mechanism of pulsating tinnitus (PT) is unclear, it's a major medical problem affecting the human health. In recent years, mechanism-related research has fallen into the bottleneck, and lacking of the targeted medical imaging quantitative information research methods is the key factor. On the basis of our team's prior achievements, we proposed a total new hypothesis that “the occurrence mechanism of PT is a multi-factor linkage of blood flow, sclerotin and conduction, and the brain-changing mechanism is the tinnitus centralization”. We plan to research through the following points: 1, Dynamically assess the vascular and bone morphology, hemodynamics and acoustic multi-dimensional information in the temporal bone region of the PT and healthy groups based on multimodal imaging technology, clarifing the role and correlation of the above information in the occurrence and conduction of PT, and exploring the abnormal blood flow in temporal bone region and the formation mechanism of the paravascular bone wall defect; 2, Then establish the imaging- morphology-blood flow-acoustic multiphysics coupling method, using multi-variable analysis and machine learning method to construct the integrated quantitative evaluation model of PT cause-treatment-prognosis, thereby multi-level verifing the PT multi-factor linkage mechanism; 3, Based on multimodal neuroimaging data, the discriminant analysis of pattern recognition is used to explore the changing characteristics of the key functional connection pathways in brain centralization, to establish the brain network feature model of PT, and to reveal the mechanism of PT brain changes. Finally we would like to establish an accurate and personalized diagnosis and treatment simulation system of PT, and provide a scientific basis and assessment method for PT risk prediction, treatment strategy and efficacy prediction.