The Personalised Nutrition 2.0 Congress in Amsterdam gave Tal Korem of the Weizmann Institute in Israel an opportunity to present his findings to an audience consisting of food and supplement manufacturers, retail chains, and government representatives.
His results that suggested personalised diets could modify elevated postprandial blood glucose and its metabolic consequences.
Blood glucose levels are increasing at an astonishing rate in the population. There are about 60 million people with diabetes in Europe, or about 10.3% of men and 9.6% of women aged 25 years and over.
Prevalence of diabetes is increasing among all ages in the region, mostly due to increases in overweight and obesity, unhealthy diet and physical inactivity.
Prevention vs. cure
The grim prospects for diabetic patients has led to this research, which hypothesized that people respond very differently to identical meals and therefore effective nutrition guidelines must be personally tailored.
The thinking focuses on the gut microbiome -- unique to each individual -- that coupled with a person’s dietary habits allows the adoption of a personalised diet to keep their blood sugar levels at a lower, safer level.
Therefore gut microbiota, blood parameters, physical activity and lifestyle behaviour data were fed into a machine-learning algorithm. The machine accurately predicted a personalised postprandial glycaemic response to real-life meals.
“Employing similar individualised prediction of nutritional effects on disease development and progression may also be valuable in rationally designing nutritional interventions in a variety of inflammatory, metabolic, and neoplastic multi-factorial disorders,” said Korem.
“More broadly, accurate personalised predictions of nutritional effects in these scenarios may be of great practical value, as they will integrate nutritional modifications more extensively into the clinical decision-making scheme,” he added.
Source: Cell
Published online ahead of print, doi.org/10.1016/j.cell.2015.11.001
“Personalized Nutrition by Prediction of Glycemic Responses.”
Authors: Zeevi et al.