Research On Improved ANN Cancer Diagnosis Method With Fusion Of Knowledge

With the development of economy,people’s pace of life is accelerated and the pressure is increased,and with the pollution and deterioration of the environment,the morbidity and mortality of various diseases especially cancer increased year by year.Cancer has always been a difficult problem in human medicine.It is a serious threat to human health and social development.The most effective way to reduce cancer mortality and treat patients effectively is to detect it earlier.Therefore,the study of the use of clinical information for cancer diagnosis to improve the accuracy and efficiency of cancer diagnosis is an important part of cancer prevention and control work.The pathological examinations in the process of cancer diagnosis seriously hurt patients’ body.Besides,the existing cancer diagnosis method based on data mining is only focus on the accuracy or interpretability of diagnostic results and the current research on cancer diagnosis is mostly discrete and reduplicative at current stage.According to these problems,this paper proposes an improved artificial neural network(ANN)cancer diagnosis method based on knowledge element.Firstly,feature selection is carried out to extract the most explanatory subset of features to improve the explanatory capability and accuracy of the model.Secondly,evolutionary computation is employed to learn the network structure and weights,with which the correlation between clinical information and prostate cancer can be identified for diagnosis of prostate cancer.And the multi-optimization method is used to optimize the structure and parameters of the model during training process,thus providing a set of effective diagnostic model to meet the different decision-making preferences of medical workers.Then,analyzed the network structure and the forming process of the obtained interpretable model to extract the comprehensible rules.After that,based on the theory of knowledge element,the knowledge representation and management of cancer diagnosis model is presented.Finally,numerical experiments were performed by using prostate cancer data from the National Center for Clinical Medical Sciences,and the scientificity and practicability of the proposed method was verified.The results show that the improved ANN cancer diagnosis method based on knowledge element in prostate cancer diagnosis is effective.It can provide multiple alternative diagnostic models for medical workers to choose based on their different decision preferences and practical needs.Feature selection based on genetic algorithm can effectively improve the performance of the neural network model.The model management method based on knowledge element can be used to express and manage the cancer diagnosis model well.