基于任务态功能近红外成像检测微意识状态患者认知功能的研究.

Objective To investigate the effect and accuracy of task state functional near infrared spectroscopy (f NIRS) in detecting advanced cognitive ability of patients with micro state of consciousness(MCS). Methods Using f NIRS and active motor imagery（ MI） paradigm,5 patients with micro state of consciousness (MCS) and 6 normal controls were tested; Support vector machine（ SVM） and linear discriminant analysis (LDA) were used to classify the responses of the subjects to the “yes” or “no” questions. Results For the basic fact “yes” problem, the response of blood oxygen in the motor area and brain frontal lobe area of the normal control increased significantly, but no significant changes of blood oxygen were detected for the basic fact “no” problem. The relative reliable response of blood oxygen similar to that of normal control was detected in 3 MCS patients in the active Mi paradigm, that is, during the “yes” task, the concentration of oxyhemoglobin (HbO2) in the motor and frontal lobe brain areas increased significantly. During the basic fact “no” task, there was no obvious response to blood oxygen. One of them was able to answer all the questions accurately by this method, and SVM was better than LDA classifier. Conclusions The f NIRS and active MI paradigm can be used to study the brain function of patients with MCS, and can be used as objective indicators to evaluate the level of consciousness and residual consciousness of patients with MCS. In addition, the “yes ” and “no ” problems of motor imagery are classified by machine learning method, which lays a foundation for further communication between patients with consciousness disorders and the outside world through brain computer interface. [ABSTRACT FROM AUTHOR]