Analysis of complications after transcatheter arterial chemoembolization based on deep learning

The aim of this exploration is to analyze the elements related to infections as well as neurological damage complications after transcatheter arterial chemoembolization (TACE) by computer information health analysis using deep learning. In this exploration, there were 80 primary liver cancer patients who were selected as the study subjects. After each patient was treated with TACE, analysis is made on their postoperative complication incidence as well as the complication differences among different groups. Moreover, comparison is carried out among the levels of alpha-fetoprotein, grades of liver function, iodized oil dosage, arteriovenous fistula, tumor blood supply, as well as tumor morphology in operation. The patient's information was input into the deep learning system to analyze the patient's disease status. There were statistical differences at P < 0.05. The research results suggested that ectopic embolization of iodized oil, myelosuppression, liver function impairment, gastrointestinal symptoms, fever, as well as infection could lead to neurological damage. The incidence that they caused was 1.25%, 22.5%, 77.5%, 86.25%, 90%, and 58.75%, respectively. There was no obvious relationship between alpha-fetoprotein and the postoperative complications mentioned above; moreover, tumor morphology as well as tumor blood supply could affect the postoperative infection, fever, gastrointestinal symptoms, as well as liver function damages obviously. In addition, the classification of liver function could greatly influence the symptoms of postoperative gastrointestinal, myelosuppression, as well as damages of liver function. Therefore, the analysis of TACE postoperative neurological damage complications as well as correlated factors revealed that it was necessary to take appropriate nursing treatment measures, which can not only enhance the TACE surgery’s success rate, but also prolong the survival of patients.