In spine surgery, nerve complications are inevitable question for clinicians. Spinal surgery needs to monitor the complete function of the spinal cord and nerve root effectively. However, single neural electrophysiological monitoring technology has limited scope. Influenced by many factors such as anesthesia, blood pressure, body temperature, position, the variant neural signals need to be monitored in real-time by clinician. In this paper, we propose an intraoperative neurological monitoring system for robot assisted minimally invasive spine surgery using electromyography. Signal preprocessing, feature extraction, and decision making procedures have been applied to achieve automatic monitoring and early warning. We also designed the monitoring software interface, which shows the feature of the signals and the reference baseline. The injury probability will also provide to physicians for decision making. Experimental results verified the proposed method.