Bioimpedance-based identification of malnutrition using fuzzy logic

Area of Science:

• Biomedical Engineering
• Clinical Nutrition
• Medical Informatics

Background:

• Protein-energy malnutrition (PEM) significantly impacts patient quality of life, hospital stay duration, and mortality rates.
• Current methods for assessing nutritional status and identifying malnutrition lack simplicity and objectivity.
• Effective nutritional assessment is crucial for improving patient outcomes and reducing healthcare costs.

Purpose of the Study:

• To develop and validate a novel assistance system for physicians to objectively assess nutritional status.
• To leverage body composition analysis derived from bioimpedance spectroscopy for nutritional status evaluation.
• To create a system capable of identifying malnourished patients through objective data comparison.

Main Methods:

• Bioimpedance spectroscopy (BIS) measurements were conducted on 688 healthy subjects and 707 patients (94 with malnutrition-causing diseases, 613 from hospital screening).
• Body composition was calculated from BIS data, and patient nutritional status was assessed using the Subjective Global Assessment (SGA) score.
• A fuzzy logic system or decision tree was employed to compare patient body composition features against healthy reference ranges, identifying malnutrition.

Main Results:

• The developed assistance system demonstrated the ability to identify malnourished patients.
• The system's core principle relies on the correlation between body composition and nutritional status.
• Validation against SGA scores confirmed the system's efficacy in nutritional status assessment.

Conclusions:

• The novel assistance system provides a simple and objective method for identifying malnourished patients.
• This system can be effectively utilized for screening and monitoring the nutritional status of hospital patients.
• Objective nutritional assessment using body composition analysis has the potential to improve patient care and outcomes.