A toolkit for quantifying individual response to herbal extracts in metabolic and inflammatory stress.

Abstract

This study developed a health assessment tool to analyze dynamic stress responses and resilience with the PhenFlex challenge. This study integrated a health space model and machine learning to quantify and visualize the impact of herbal extracts on inflammatory and metabolic health at the individual level. Two randomized, double-blind, placebo-controlled crossover trials were conducted involving participants with PhenFlex challenge after overnight fasting. Blood samples were collected, and a machine learning algorithm was used to predict health estimation scores based on metabolic and inflammatory responses. The resulting health space model visually represents individuals' health status in a 2-D space. Intervention with herbal extracts (e.g., Angelica keiskei, AK, and Capsosiphon fulvescens, CF) resulted in higher health scores in the health space, indicating improved health. This research emphasizes the quantification of phenotypic changes for personalized nutrition and health optimization. Overall, this study provides a valuable toolkit for validating herbal extract efficacy and extends its application to personalized nutrition.