腦中風是台灣重大疾病之一，而中風後的復健十分重要，有效率的復健策略可以提高中風病人痊癒的可能性，因此對於中風後的預測與判斷更是重要的一環。許多關於個體差異的神經影像學研究都注重於在建立大腦測量值與特徵（例如智力，記憶力，注意力或疾病症狀）之間的相關關係。藉由科技的進步及大量收集的數據資料能夠建立一個能從神經影像取得的大腦關聯性來預測中風病人復健狀況的模型。本論文利用聯結體來建立模型，這個方法主要分成四個部分，特徵選取、特徵總結、建立模型及評估預測效果。本論文運用中風病人的靜息態腦功能性磁振造影及復健量表參數來完成聯結體模型的建立，以利於往後復健策略的改善與調整。;Stroke is one of the major diseases in Taiwan, and rehabilitation after a stroke is very important. Effective rehabilitation strategies can increase the possibility of stroke patients′ recovery. Therefore, it is important to predict and judge after stroke. Many neuroimaging studies of individual differences have focused on establishing correlational relationships between brain measurements and traits such as intelligence, memory, and attention, or disease symptoms. Through the advancement of technology and a large amount of collected data, a model can be established to predict the rehabilitation status of stroke patients by brain correlation obtained from neuroimaging. We will present Connectome-based predictive modeling in this case. This method is mainly divided into four parts, feature selection, feature summarization, model building and assessment of prediction significance. We use fMRI and rehabilitation scale parameters of stroke patients to complete the establishment of the CPM model, in order to facilitate the improvement and adjustment of rehabilitation strategies in the future.
