Design and Application of Intelligent Medical Care Beds Empowered by EEG Technology

Abstract

With the continuous development of domestic brain-computer interface technology, in response to clinical pain points such as high nursing labor consumption and high risk of pressure ulcers for long-term bedridden patients, this paper designs an Intelligent medical bed system empowered by EEG. The system adopts an edge intelligent decision-making architecture composed of embedded control and brain electroencephalogram (EEG) rehabilitation upper-level modules. At the control level, non-invasive electrodes are used to collect single-channel EEG signals from the frontal lobe, and an OptimizedCNN deep learning model is employed to decode the patient's movement intentions of 'resting/left turn/right turn' in real-time, driving the bed's segmented motors to achieve closed-loop turning control. At the nursing assistant level, a catheter length prediction model is built based on BMI and height measurement data to assist with precise PICC placement. Simulation test results show that the system's accuracy in recognizing turning intentions reaches 68.0%. Combined with a triple safety fuse mechanism, it significantly improves nursing efficiency and patient safety, achieving a shift from experience-based care to data-driven and intelligent nursing.