Role of artificial intelligence in predicting disease-related malnutrition - A narrative review

Area of Science:

• Clinical Nutrition
• Medical Informatics
• Artificial Intelligence in Healthcare

Background:

• Disease-related malnutrition (DRM) impacts 30-50% of hospitalized patients, leading to increased complications and healthcare costs.
• Current manual methods for DRM detection are often inaccurate and inefficient, contributing to underdiagnosis.
• There is a critical need for advanced tools to improve the identification and management of DRM.

Purpose of the Study:

• To review the application of artificial intelligence (AI), machine learning (ML), and deep learning (DL) in predicting and managing DRM.
• To assess the clinical utility, benefits, and drawbacks of various AI models for DRM.
• To explore the integration of AI tools with electronic health records for real-time patient management.

Main Methods:

• A narrative review of existing literature on ML and DL models for DRM.
• Examination of the clinical applicability, advantages, and limitations of these AI models.
• Analysis of how AI integration into electronic health records can automate risk detection and optimize management.

Main Results:

• ML and DL models demonstrate significant potential for accurate nutritional status assessment and complication prediction in DRM patients.
• AI facilitates enhanced clinical decision-making and more efficient resource allocation.
• Implementation challenges include the need for large, standardized datasets and seamless integration with existing healthcare systems.

Conclusions:

• AI presents a promising avenue for proactive DRM management in clinical settings.
• Interdisciplinary collaboration is essential to address implementation barriers and maximize AI's benefits for patient care.
• AI-driven solutions can lead to improved patient outcomes and reduced healthcare burdens associated with DRM.