Development of NANDA nursing diagnoses in gynecological cancer patients via machine learning and data mining methods

This study was conducted to develop nursing diagnoses in gynecological patients, which are included in the NANDA care plans, which are the most important guides for nurses, by using machine learning and data mining methods. A total of 304 patients who had been diagnosed with a gyneco-oncological disorder in Pamukkale University Hospital between 2015 and 2021 and had undergone or had been scheduled for surgery were included in the study. The dates when patients stayed in the hospital were divided into preoperative and postoperative periods. Patients' blood tests done during this process, presenting complaints, pathology reports, gyneco-oncological diagnoses, chronic diseases, and surgery information were obtained from the hospital information management system. Factors that may affect nursing diagnoses in the data were investigated based on the relevant literature. Following this review, primary distinguishing factors were selected from patient files to be integrated as input variables. After the selection of appropriate artificial intelligence software, the data cleaning and transformation procedures were conducted. Necessary trials were performed to determine the most suitable algorithm, and Multilayer Perceptron and J48 yza databases were selected. As a result of the study, 17 nursing diagnoses were determined. Relevant data that affected nursing diagnoses were identified. The average accuracy rate of these diagnoses was 98%. The collection of data from educated nurses will lead to the emergence of healthier nursing diagnoses for machine learning and data mining methods in future studies. In addition, as a result of the increase in studies conducted with this method, health professionals will be able to take an active role in the diagnosis, treatment, and care of their patients only with the data analyzed. Health professionals should be encouraged to use new methods related to technological developments. Keywords: NANDA nursing diagnosis, machine learning, artificial intelligence, data mining