An Optimized Multilayer Perceptron Model for Disease Risk Prediction in Psychiatric and Psychological Nursing

Abstract

In psychiatric and psychological nursing, the early identification of high-risk individuals is essential for implementing timely and personalized interventions. This paper presents a disease risk prediction model based on a Multilayer Perceptron (MLP) neural network, designed to classify patients into high-risk and low-risk categories for multiple diseases. The model leverages patient data, including age, gender, medical history, and medication usage, to perform multi-label classification, effectively capturing complex nonlinear relationships between input features and disease risks. Data preprocessing techniques, such as handling missing values, normalization, and feature selection, were applied to improve model performance. The model was trained using the Adam optimizer to ensure rapid convergence and robust learning, while L2 regularization and mini-batch gradient descent were employed to enhance generalization and prevent overfitting. Experimental results indicate that the MLP model significantly outperforms traditional methods like logistic regression and random forest in terms of accuracy, precision, recall, and F1-score. This study offers an advanced tool for risk assessment in psychiatric nursing, providing a foundation for more accurate and personalized patient care interventions.

Ähnliche Arbeiten