Based on Natural Intelligence and Artificial Intelligence Techniques to Determine both Carbohydrates and Sugar Intake Amount and PPG Predictions Using Viscoplastic Energy Model of GH-Method: Math-Physical Medicine (No. 1030, Viscoelastic Medicine Theory #428)

Abstract

Gerald C. Hsu

The molecular structures of plants and animals display a variety of colors with different color shades, each corresponding to a unique optical wave with distinct frequency, amplitude, and wavelength. To capture these nuances, the author employs a 12-megapixel camera on his Apple iPhone 13, allowing for high-resolution imaging. This rich dataset enables precise differentiation of color shades, thanks to the iPhone's computational power and user convenience.

Utilizing this capability, the author has developed a unique mathematical algorithm that maps these 12-megapixel color data to the three fundamental properties of color waves of food and meals, facilitating the analysis of consumed food's internal makeup, such as carbohydrate and sugar content—vital information for managing his diabetes. This system, dubbed "Carbs AI" (artificial intelligence), is enhanced by the author's self-learned extensive knowledge of food nutrition, acquired over past 15 years and referred to as "Carbs NI" (natural intelligence). The method calculates postprandial plasma glucose (PPG) levels by combining linear elasticity theory with fasting plasma glucose (FPG) and post-meal step count (Steps), two other significant factors affecting PPG.

Drawing on his personal data collected within past 10 years from May 1, 2015, to February 5, 2024, the author explores PPG prediction via AI and NI, integrating three inputs from both Carbs AI and NI, FPG, and Steps. This study also evaluates the prediction accuracy and curve correlation between AI and NI models with their vital energy distributions information using the space-domain viscoplastic theory and energy model (SD- VMT).

PDF