The latest evidence indicates that persistent concentrated urine may trigger chronic metabolic and kidney diseases. Recent results suggest that a daily urinary concentration of 500 mOsm/kg indicates optimal hydration. 

The present study provided personalized advice for daily water intake considering personal intrinsic (age, sex, height, weight) and extrinsic (food and fluid intakes) characteristics to acquire a target urine osmolality (UOsm) of 500 mOsm/kg using machine learning and optimization algorithms. It analyzed data from clinical trials on hydration (four randomized and three non-randomized trials) and tested several machine learning methods to predict UOsm. It then evaluated the predictive performance of the developed algorithm against current dietary guidelines. 

The study found features linked to urine production and fluid consumption as the most important, with relative importance values ranging from 0.10 to 0.95. XGBoost occurred as the most performing approach (Mean Absolute Error (MAE) = 124.99) to predict UOsm. The developed algorithm displayed the highest overall correct classification rate (85.5%) than that of dietary guidelines (77.8%). 

This machine-learning application delivers personalized advice for daily water intake to achieve optimal hydration and can be considered a primary prevention tool to balance the increased incidence of chronic metabolic and kidney diseases.

Dolci A, Vanhaecke T, Qiu J, et al. Personalized prediction of optimal water intake in the adult population by blended use of machine learning and clinical data. Sci Rep.2022; 12. https://doi.org/10.1038/s41598-022-21869-y