Science
Diagnose before prescribe.
MetaAge is a deep learning biological aging clock. It translates routine blood parameters into a structured estimate of biological age, built on peer-reviewed research and a proprietary Holistica World study, and designed to support clinical conversations rather than replace them.
Methodology
How an estimate is built.
Input
Blood parameter values with units, plus age and sex.
Normalisation
Values are checked, unit-harmonised and placed in context.
System modelling
A deep learning model reads patterns across markers, groups them into body systems and scores each system.
Report
Systems combine into overall biological age and pace of aging.
Philosophy
Two traditions, one question.
Western clinical diagnostics brings precise measurement. Indian Vedic science brings a whole-person view of balance and constitution. Holistica World draws on both, asking what is truly happening in the body before choosing what to do next.
Deep learning
Model class
Routine blood
Inputs
6
Studies referenced
Holistica World
Proprietary study
Evidence base
The research MetaAge builds on.
Six independent studies show that deep learning and related models can estimate biological age from routine clinical data, and that the gap between biological and chronological age tracks disease, hospitalization and mortality. They are the scientific foundation MetaAge builds on. They are not a validation of the MetaAge model itself, which is the subject of our proprietary study.
Cell Reports, 2022
Distinct biological ages of organs and systems
In 4,066 adults aged 20 to 45 in Shenzhen, 403 blood, stool, skin and fitness features were grouped into nine categories to build an aging-rate index per organ system. Different systems aged at different rates, shaped by genetics and lifestyle.
Relevance to MetaAge: Supports MetaAge's system-level scoring: one person carries several biological ages, not one.
Nature Aging, 2024
A hematology-based clock derived from the Study of Longitudinal Aging in Mice
A deep neural network trained on routine blood counts and two metabolic indices from 2,562 mice predicted blood age. The gap between predicted and chronological age was associated with mortality risk and frailty, and platelets were the strongest predictor.
Relevance to MetaAge: Preclinical evidence that deep learning on routine blood data produces an aging gap with biological meaning. Mouse data, so it supports the method rather than any human result.
Scientific Reports, 2025
A sex-adjusted 7-biomarker clinical aging clock for translational preventative medicine
A deep learning model trained on 59,741 healthy blood samples from a Southeast Asian hospital cohort was reduced to six routine biomarkers plus sex. After a skew correction it was validated in NHANES and UK Biobank, where the age gap was raised in diabetes and chronic kidney disease, appeared up to five years before diagnosis and predicted mortality.
Relevance to MetaAge: A close analogue to MetaAge's approach: deep learning on routine bloods, trained on a non-Western population and tested across ethnicities.
bioRxiv preprint, 2025
A combined blood and biometric aging clock built on NHANES
A gradient boosting model trained on 5,542 healthy NHANES participants combines blood tests with blood pressure, BMI and waist circumference to keep testing costs low. Glycated hemoglobin was the most influential input, and fast agers had higher odds of hypertension, obesity and heart attack.
Relevance to MetaAge: Shows how affordable inputs and transparent feature importance make an aging clock practical in routine clinics. Preprint, not yet peer reviewed, and a machine learning rather than deep learning model.
GeroScience, 2024
Blood biological age and physical activity during rehabilitation in geriatric inpatients (RESORT)
In 111 geriatric rehabilitation inpatients in Melbourne, blood biological age was estimated with a hematological aging clock. Each additional year of biological age was associated with higher odds of high sedentary time and trended towards less physical activity.
Relevance to MetaAge: Shows a blood-based aging clock read against real functional behavior in a clinical setting, with potential use in rehabilitation planning.
Aging and Disease, 2026
Difference between biological age and chronological age predicts mortality and hospitalization
In 2,597 adults at Sheba Medical Center followed for a mean of 9.2 years, a deep learning model on routine laboratory tests estimated biological age. Each extra year of age gap carried a 15% higher mortality hazard and a 6% higher hospitalization rate, and a gap of 3 years or more marked clearly higher risk.
Relevance to MetaAge: The strongest outcome evidence for the age-gap concept from routine blood tests. It used a third-party model, so it supports the approach rather than MetaAge specifically.
Our own cohort, performance metrics and publication route will be added here as the proprietary study is finalized.
Responsible use
Insight that supports clinical judgment.
MetaAge provides estimates, not diagnoses. It is not intended to detect, diagnose, treat, cure or prevent any disease. Results should be interpreted by a qualified professional within the full clinical context, including history, examination and other investigations. No treatment decision should be based on a MetaAge result alone.
For clinics, centers and platforms
Bring MetaAge into your practice.
Offer a branded biological age report built on bloodwork you already run.