The study describes a real-time neuro-gaming system that uses neural signals to make games more accessible and has great promise for helping quadriplegic patients recover. Users can engage with computers and obtain visual feedback through video games by visualizing physical activities; this promotes mental workouts that increase neuroplasticity and brain activation. Using both long short-term memory (LSTM) and convolutional neural network (CNN) models, the system uses sophisticated deep learning techniques to categorize time-series electroencephalography (EEG) data into three different classes with an accuracy of 71.5% and 66%, respectively. Preprocessed EEG signals are integrated into the gaming input system to provide accurate and responsive real-time game control. Furthermore, bio-signals are used in a real-time video game created with the pygame library. A neural network trained on EEG signals from players completing mental tasks is used for gaming in a test using a g.Nautilus headset, the system demonstrated the interactive and therapeutic potential of neuro-gaming by successfully manipulating a BCI-controlled avatar with 71.5% accuracy.