A Deep Learning-Based Text Classification of Adverse Nursing Events.

Area of Science:

• Healthcare Management
• Medical Informatics
• Artificial Intelligence in Medicine

Background:

• Adverse nursing events negatively impact patient outcomes, increase healthcare costs, and disrupt medical operations.
• Effective management of adverse nursing events is crucial for patient safety and healthcare system development.
• Current manual classification of unstructured nursing event reports is inefficient and yields inaccurate data.

Purpose of the Study:

• To evaluate deep learning-based classification methods for adverse nursing events in healthcare systems.
• To propose and validate a novel text classification model for adverse nursing events.

Main Methods:

• Extensive evaluation of various deep learning-based classification techniques.
• Development and implementation of a proposed deep learning text classification model.
• Comparative analysis of the proposed model against existing methods using experimental data.

Main Results:

• The proposed deep learning model demonstrated exceptional performance in classifying adverse nursing events.
• The model achieved superior results across various evaluation metrics compared to traditional methods.
• Deep learning approaches offer a viable solution for analyzing unstructured healthcare data.

Conclusions:

• Classifying adverse nursing events using deep learning enhances the accuracy and efficiency of data analysis.
• The proposed model provides a significant advancement in managing and reducing adverse nursing events.
• Implementing advanced text classification methods is essential for improving patient safety and healthcare quality.