关键词: Cluster analysis Decision trees Metabotypes Personalised nutrition Targeted nutrition

Mesh : Adult Blood Glucose / analysis Blood Pressure Cholesterol / blood Cholesterol, HDL / blood Cluster Analysis Diet Female Humans Inflammation / blood Male Metabolomics / methods Middle Aged Nutrition Surveys Nutritional Physiological Phenomena Nutritional Status Precision Medicine / methods Triglycerides / blood

来  源:   DOI:10.1002/mnfr.201400591   PDF(Sci-hub)

Abstract:
METHODS: Personalised nutrition can be defined as dietary advice that is tailored to an individual. In recent years, the concept of targeted nutrition has evolved, which involves delivering specific dietary advice to a group of phenotypically similar individuals or metabotypes. This study examined whether cluster analysis could be used to define metabotypes and developed a strategy for the delivery of targeted dietary advice.
RESULTS: K-means clustering was employed to identify clusters based on four markers of metabolic health (triacylglycerols, total cholesterol, direct HDL cholesterol and glucose) (n = 896) using data from the National Adult Nutrition Survey. A decision tree approach was developed for the delivery of targeted dietary advice per cluster based on biochemical characteristics, anthropometry and blood pressure. The appropriateness of the advice was tested by comparison with individualised dietary advice manually compiled (n = 99). A mean match of 89.1% between the methods was demonstrated with a 100% match for two-thirds of participants.
CONCLUSIONS: Good agreement was found between the individualised and targeted methods demonstrating the ability of this framework to deliver targeted dietary advice. This approach has the potential to be a fast and novel method for the delivery of targeted nutrition in clinical settings.
摘要:
暂无翻译
公众号