Brain-Computer Interface (BCI) systems have recently emerged as a prominent technology for assisting paralyzed people. Recovery from paralysis in most patients using the existing BCI-based assistive devices is hindered due to the lack of training and proper supervision. The system's continuous usage results in mental fatigue, owing to a higher user concentration required to execute the mental commands. Moreover, the false-positive rate and lack of constant control of the BCI systems result in user frustration. The proposed framework integrates BCI with a deep learning network in an efficient manner to reduce mental fatigue and frustration. The Deep learning Brain Network (DBN) recognizes the patient's intention for upper limb movement by a deep learning model based on the features extracted during training. DBN correlates and maps the different Electroencephalogram (EEG) patterns of healthy subjects with the identified pattern's upper limb movement. The stroke-affected muscles of the paralyzed are then activated using the obtained superior pattern. The implemented DBN consisting of four healthy subjects and a quadriplegic patient achieved 94% accuracy for various patient movement intentions. The results show that DBN is an excellent tool for providing rehabilitation, and it delivers sustained assistance, even in the absence of caregivers.

脑机接口（BCI）系统最近已经成为帮助瘫痪者的重要技术。由于缺乏培训和适当的监督，使用现有的基于BCI的辅助设备使大多数患者的瘫痪恢复受到阻碍。由于执行精神命令需要更高的用户专注度，因此系统的连续使用会导致精神疲劳。而且，假阳性率和对BCI系统的持续控制不足导致用户感到沮丧。拟议的框架以有效的方式将BCI与深度学习网络集成在一起，以减少精神疲劳和挫败感。深度学习脑网络（DBN）通过基于在训练过程中提取的特征的深度学习模型来识别患者上肢运动的意图。DBN将健康受试者的不同脑电图（EEG）模式与确定的模式的上肢运动相关联并映射。然后，使用获得的上乘模式激活瘫痪者的受中风影响的肌肉。由四个健康受试者和一个四肢瘫痪患者组成的已实施DBN达到了针对各种患者运动意图的94％的准确性。结果表明，DBN是提供康复的出色工具，即使在没有护理人员的情况下，它也可以提供持续的帮助。由四个健康受试者和一个四肢瘫痪患者组成的已实施DBN达到了针对各种患者运动意图的94％的准确性。结果表明，DBN是提供康复的出色工具，即使在没有护理人员的情况下，它也可以提供持续的帮助。由四个健康受试者和一个四肢瘫痪患者组成的已实施DBN达到了针对各种患者运动意图的94％的准确性。结果表明，DBN是提供康复的出色工具，即使在没有护理人员的情况下，它也可以提供持续的帮助。