The Role of Machine Learning in LOS Reduction for Patients Affected by Lower Limb Fracture

Area of Science:

• Orthopedic surgery
• Medical informatics
• Artificial Intelligence

Background:

• Length of Stay (LOS) estimation is vital for hospital efficiency and patient care.
• Accurate LOS prediction is particularly important for orthopedic trauma patients, especially those with lower limb fractures.
• Optimizing resource allocation and patient management relies on effective LOS forecasting.

Purpose of the Study:

• To analyze and predict the Length of Stay (LOS) for patients with lower limb fractures.
• To implement and evaluate five neural network-based classifiers for LOS prediction.
• To compare the performance of these neural networks against established Artificial Intelligence (AI) models from previous research.

Main Methods:

• Utilized five distinct neural network-based classifiers.
• Applied these models to predict LOS for patients with lower limb fractures.
• Compared the predictive performance of the neural networks with prior AI models.

Main Results:

• The study successfully implemented and tested five neural network classifiers for LOS prediction in orthopedic trauma patients.
• Performance metrics were generated for each neural network model.
• Comparative analysis was conducted against previously developed AI models.

Conclusions:

• Neural network-based classifiers show promise for predicting Length of Stay (LOS) in patients with lower limb fractures.
• The findings contribute to optimizing hospital resource management and patient care strategies.
• Further research can refine these AI models for enhanced clinical decision-making.