Study on Compensation of Blood Oxygen Sensor’s Ambient Light Based on GA-BP Neural Network

Abstract In order to reduce the interference of ambient light on near-infrared blood oxygen sensors (used measure saturation human tissue or circulating during Extracorporeal membrane oxygenation treatment) and improve measurement accuracy a sensor, compensation method is proposed. This does not need an sensor additional circuit. The intensity calculated by changing value emitted LED in combining received sensor. Genetic Algorithms-Back Propagation(GA-BP) neural network modeled, detection evaluation values are taken as input characteristics, gas analysis true expected output, model used for prediction. results show that this higher than traditional fitting results, error between prediction basically within 5%.