意识障碍疾病的个体化脑功能网络分析技术研究

Severe brain injury can lead to disorders of consciousness (DOC). To date, the diagnostics and prognostics for a DOC patient are based on physician observation of the patient's behavior, which could bring about high misjudgment rate. To provide more accurate and objective information for the prognosis of the DOC, we developed a computational model that combines resting state functional MRI with clinical characteristics to predict one year outcome at the single-patient level. The model discriminated between the patients who would later recover consciousness and those who would not with an accuracy of around 90% on three datasets from two medical centers. To our knowledge, this is the current state-of-the-art accuracy to predict the probability of recover consciousness for chronic DOC patients. Despite this, a much larger cohort will be needed for further validation and our analysis method requires improvement. One of the most concern is that the present method is based on the classical region of interest (ROI) analysis, which cannot fully explore individual differences in brain structure and function. Therefore, we propose to develop a novel brain functional network analysis method for DOC patients, which includes the following four research contents, "mapping functional networks in individual healthy brains", "measuring brain structural damages in DOC patients", "detecting brain functional networks for individual DOC patients", and "modeling classification and prognostication models". We believe that these advantages could help to check the patient's residual brain function, and offer an objective and quantitative information for the diagnosis and prognosis of DOC patients.

严重脑损伤导致的意识障碍尚无有效诊断和预后预测方法，仅凭临床量表误诊率高达40%，给家庭和社会带来沉重负担。申请人和合作单位已初步建立了一个基于脑影像和临床指标的预测模型，用于检测病人的残余脑功能，预测其1年后是否恢复意识。我们在北京和广州两个医疗中心的三套数据集上取得了目前公开报道的最优的预测准确率。尽管如此，该模型仍然需要多中心大样本的验证和方法学上的改进。其中最突出的一个问题是：目前的影像分析方法没有充分考虑病人脑功能网络的个体化差异。本项目将从人脑“结构”和“功能”个体差异两个角度，从“正常被试个体脑功能网络计算方法的研究”、“意识障碍病人脑结构损伤的检测”、“意识障碍病人脑功能网络个体化分析”、以及“意识障碍辅助诊断与预后预测模型的设计”等四个方面进行研究，建立意识障碍病人个体脑功能网络的计算方法，为意识障碍疾病的诊断、预后判断以及疗效评估提供客观依据。

意识障碍患者俗称“植物人”，病人长期处于“死不了，治不好”的艰难境地，给家庭和社会带来巨大精神痛苦和沉重经济负担。对病人意识恢复可能的预测，将直接影响临床治疗策略的选择，甚至是亲属对病人生死去留的抉择。但是，目前临床对病人的预后判断完全依赖医生根据临床观察和个人经验做出，主观性较强，易受多种因素干扰，对预后的误判可能导致严重后果。通过本项目四年的工作，我们圆满完成了各项计划任务。具体包括，我们创新性地提出了“任务EEG检测结合静息fMRI检测是提高意识障碍脑功能检测临床效率的优化策略”，并优化改进了基于静息功能磁共振影像和临床信息的意识障碍预后预测模型，发布了开源的软件包，在首都医科大学附属北京天坛医院、南方医科大学珠江医院、中南大学湘雅医院等国内多家医院意识障碍患者诊疗中进行应用，辅助医生进行患者预后的评估工作。.在本项目资助下，发表标注资助号的国际期刊论文6篇，申请国家发明专利1项。本项目提出的预后预测模型被著名医学专家作为“年度研究亮点”重点推介，相关研究结果写入意识障碍临床专家共识（2020年），并获得2021年北京医学科技奖二等奖。