Nursing Value Analysis and Risk Assessment of Acute Gastrointestinal Bleeding Using Multiagent Reinforcement Learning...

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Clinical Risk Management

Background:

• Gastrointestinal bleeding (GIB) poses significant health risks, necessitating accurate nursing value analysis and patient risk assessment.
• Current risk assessment techniques for GIB patients are inconsistent, highlighting a need for improved methodologies.
• Machine learning (ML) offers potential for enhancing risk evaluation in clinical settings.

Purpose of the Study:

• To develop and evaluate a unique machine learning-based framework for nursing value analysis and risk assessment in GIB patients.
• To construct a predictive model for hospital-based interventions or mortality risk in GIB patients.
• To compare the performance of the proposed ML model against existing rating systems.

Main Methods:

• Dataset collection and preprocessing followed by feature extraction using local binary patterns (LBP).
• Classification was performed using a fuzzy support vector machine (FSVM) classifier.
• A multiagent reinforcement learning algorithm, optimized with the spider monkey optimization (SMO) algorithm, was employed for risk assessment and nursing value analysis.

Main Results:

• The proposed machine learning framework demonstrated strong prognostic efficacy in individuals with GIB.
• Performance metrics including classification accuracy, AUROC, AUC, sensitivity, specificity, and precision were analyzed.
• The ML-based approach significantly outperformed traditional models in risk assessment for GIB patients.

Conclusions:

• The developed machine learning framework provides a robust and effective tool for nursing value analysis and risk assessment in GIB patients.
• The study highlights the superiority of the proposed ML method over traditional approaches for predicting adverse outcomes in GIB.
• This research paves the way for more consistent and accurate risk stratification in GIB patient care through advanced AI techniques.