Design And Implementation Of Cervical Cancer Diagnosis And Treatment Assistant System Based On Deep Learning

Investigation and research show that incidence rate and mortality rate of cervical cancer are increasing in recent years,especially among women living in areas with poor economic development.On the one hand,in the diagnosis and treatment of cervical cancer,doctors’ behavior is not standardized.Although there are experts to develop the guidance of cervical cancer clinical pathway specification,affected by the dynamic evolution of the disease,the difference of medical equipment in different medical institutions and other factors,the path out rate of clinical pathway is very high.It is difficult to achieve the goal of standardizing doctors’ diagnosis and treatment behavior.On the other hand,the diagnosis and treatment process of cervical cancer is complicated.The workload of doctors in the process of making diagnosis and treatment plan for patients is large,and many factors need to be considered,which leads to long hospital stay and high cost.In this paper,aiming at the problems existing in the diagnosis and treatment of cervical cancer,this paper adopts the mature deep learning technology to propose an innovative clinical events prediction model with standardized judgment.It is based on the data in patients’ electronic medical records.Because of semi-sequential character of hospitalization data,it is pooled.And then the data is expressed in the form of embedded vector by skip-gram algorithm.The bisecting K-means mining model based on cosine similarity is established to mine the standardized diagnosis and treatment mode,which provides the measurement standard for the standardized judgment of diagnosis and treatment process.Based on positive and negative bidirectional recurrent neural network of gated recurrent unit,we construct the clinical prediction model to predict the clinical events in the next stage.Before the prediction,the model makes a standardized judgment on the diagnosis and treatment process.The comparative analysis shows that the mean average precision value of the model is improved by 5.34%compared with the case that the diagnosis and treatment process is not standardized.Based on the standard diagnosis and treatment mode mining model and clinical events prediction model,we design and implement cervical cancer diagnosis and treatment assistant system.On the one hand,the purpose is to judge whether the doctors’ past diagnosis and treatment behavior is standardized.On the other hand,the system predicts the clinical events in the next stage based on the past standardized diagnosis and treatment process,so as to assist the attending doctor to formulate a standardized diagnosis and treatment plan,and further achieve the objective of auxiliary diagnosis and treatment.The system realizes information input,information retrieval,data statistics,intelligent analysis,and other functions.In the process of diagnosis and treatment of cervical cancer patients,the system standardizes the diagnosis and treatment behavior of doctors,improve the work efficiency of doctors,shorten the hospitalization time of patients,and reduce hospitalization expenses.