Recommending healthy meal plans by optimising nature-inspired many-objective diet problem

Area of Science:

• Nutritional Science
• Computational Optimization
• Health Informatics

Background:

• Healthy eating is a global concern, amplified by the COVID-19 pandemic.
• Traditional diet plans often lack realism and sustainability due to restrictions.
• Personalized nutrition is crucial for public health and well-being.

Purpose of the Study:

• To extend the classical diet problem with realistic modeling and a novel solution approach.
• To develop a sustainable diet planning system using cooked dishes.
• To create personalized daily menus adaptable to individual needs and health conditions.

Main Methods:

• Re-modeling the diet problem as a many-objective optimization problem, inspired by animal hunting strategies.
• Utilizing a dataset of cooked dishes for realistic meal planning.
• Employing a many-objective evolutionary algorithm for problem-solving.

Main Results:

• The developed approach successfully optimizes specialized daily menus tailored to user preferences, age, gender, and body mass index.
• The system demonstrates adaptability for users with specific health issues by incorporating additional constraints and objectives.
• The optimized diet plans are realistic and sustainable, considering user expectations without undue restrictions.

Conclusions:

• This novel approach provides a flexible and effective decision support mechanism for both individuals and dietitians.
• The methodology offers a pathway to personalized and sustainable healthy eating habits.
• The system's adaptability highlights its potential for broader application in public health and clinical nutrition.