A SEMANTIC DESCRIPTION MODEL OF LUNG CANCER CHROMATIC IMAGES BASED ON BAYESIAN LEARNING

By introducing the Bayesian framework to the lung cancer diagnosing problem,a semantic description model of lung cancer chromatic images is proposed, which is composed of raw image layer (RIL), image feature layer (IFL), semantic knowledge layer (SKL), and a semantic description algorithm SDA. Based on this model, a lung cancer identification system is successfully implemented.The experiment results also show that this model is very effective to identify different kinds of lung cancer at high correct rates.