Back to search

Institutes outside Greece

Artificial intelligence–powered nutrition software for healthy individuals, patients, athletes, children, elderly, and pregnant/lactating women: implementation in prevention, therapy, and optimization across primary health care, public health, and research

Abstract

dc:description

Aim: This study aimed to design, develop, and conduct a preliminary evaluation of an innovative artificial intelligence (AI) framework for generating personalized seven-day meal plans that addresses limitations commonly observed in existing dietary planning systems, including insufficient personalization, limited variety, and nutritional imbalances. Methodology: The study methodology comprised: (a) a systematic literature review to identify and analyze existing dietary software applications, (b) the definition of the system's functional requirements, (c) the development of a prototype utilizing reinforcement learning algorithms, (d) verification of system functionality, and (e) evaluation of the generated meal plans by 37 potential users and 8 registered dietitians using a five-point Likert-scale questionnaire. Statistical power was determined a priori through power analysis. Results: The proposed system successfully generated complete seven-day meal plans with an appropriate structure consisting of five meals per day. Weekly evaluation of nutritional targets demonstrated improved consistency compared to daily assessments, with mean deviations of 1.72% for energy intake, 1.80% for protein, 4.13% for carbohydrates, and 2.80% for fats. Potential users rated the AI-generated meal plans significantly higher in practical aspects such as variety, food availability, and cost-effectiveness. However, meal plans designed by dietitians achieved superior ratings in nutritional adequacy and overall acceptance. Expert dietitians identified no statistically significant differences between AI-generated and dietitian-designed meal plans across most evaluation criteria. Conclusions: The findings indicate that the proposed AI system can effectively manage the quantitative dimensions of dietary planning in a systematic manner. Nevertheless, human expertise remains essential for addressing the psychological, cultural, and holistic aspects of dietary counseling. Future research should focus on integrating multimodal data and conducting large-scale clinical studies to support the clinical implementation and validation of the system.

Degree

thesis:*
Grantor dc:publisher
Institutes outside Greece
Year dc:date
2026

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Nikitidis, Nikolaos
  • Νικητίδης, Νικόλαος

Subjects

dc:subject × 32

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/62574

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Nikitidis, Nikolaos; Νικητίδης, Νικόλαος. Artificial intelligence–powered nutrition software for healthy individuals, patients, athletes, children, elderly, and pregnant/lactating women: implementation in prevention, therapy, and optimization across primary health care, public health, and research. Institutes outside Greece, 2026. http://hdl.handle.net/10442/hedi/62574