Close Regenerative Index · Methodology

Not a model.
A decade of clinical truth
compressed into four engines.

The CRI was built because regenerative medicine in Southeast Asia operates without accountability. We changed that — one biomarker, one patient, one outcome at a time.

100K+
Blood Panels Analysed
10
Years Clinical Data
4
Proprietary Engines
1st
Clinical CRI System in SEA
Johns Hopkins University
Biostatistics · Population Analytics
MIT · Sloan
AI in Healthcare & Business Strategy
Memorial Sloan Kettering
Oncology Biomarker Intelligence
18 Years
Physical Medicine · Regenerative Practice
USP797 Compliant
Compounding · Cryogenic Laboratory, Bali
"In regenerative medicine across Southeast Asia, almost any claim can be made without clinical consequence."

We have operated in this space long enough to witness harm caused not by malice, but by the absence of measurable standards. Clinics promise outcomes. Patients pay. Neither party has a framework to determine whether biology is actually responding.

The CRI was not created as a product. It was created as a corrective. A proprietary scoring architecture that replaces opinion with evidence, replaces anecdote with longitudinal data, and replaces financial incentive with biological accountability.

If we cannot demonstrate measurable progress toward a defined biological endpoint, we do not treat. That is not a policy. That is the engine itself executing its decision logic.

From Excel pivot tables
to a clinical intelligence system.

The CRI did not emerge from a whiteboard. It was distilled from a decade of raw clinical data — initially tracked in spreadsheets, then in statistical models, then formalised through graduate-level biostatistics and AI methodology training.

2015
Raw Data Accumulation
Blood panel collection begins. Excel pivot tables. Manual statistical cross-referencing of biomarkers against outcomes across physical medicine cases.
2017
Biostatistical Framework
Johns Hopkins biostatistics coursework formalises the population analytics approach. Bayesian confidence modelling introduced to the dataset.
2019
AI Methodology
MIT Sloan and Memorial Sloan Kettering AI in healthcare training. Clinical machine learning frameworks applied to the longitudinal dataset for the first time.
2022
Engine Architecture
The four core engines are formalised. 10-biomarker minimum viable panel established. The CRI composite scoring framework enters clinical use at Dripdok.
2025
First in SEA
CRI becomes the first clinically deployed regenerative medicine accountability index in Southeast Asia. Wearable integration and RVI scoring launched.

The Four Engines of the Close Regenerative Index

Each engine addresses a distinct failure mode in conventional regenerative medicine assessment. Together they produce a composite biological score — the CRI — that is objective, reproducible, and clinically defensible.
Engine 01
Active
Biomarker Convergence Engine
BCE · Multi-Axis Panel Scoring

Synthesises a minimum of 10 biomarkers across metabolic, inflammatory, hormonal, and cellular health axes into a single convergent score. Unlike standard reference-range interpretation, the BCE evaluates biomarker relationships — identifying clinical patterns that individual values cannot reveal. Derived from population analytics across 100,000+ panels.

CRPNAD+HbA1cIGF-1Telomere proxyInflammatory axis
Engine 02
Active
Regenerative Velocity Index
RVI · Temporal Rate Analysis

Measures the rate of biological change rather than static values at a single point in time. The RVI computes trajectory across serial panels, distinguishing genuine regenerative response from assay noise or placebo drift. This is the engine that determines whether a protocol is actually working — and how fast.

Serial Delta scoringBayesian CIVelocity thresholdProtocol response
Engine 03
Active
Bayesian Confidence Calibrator
BCC · Probabilistic Uncertainty Modelling

Every clinical biomarker carries measurement uncertainty. The BCC applies Bayesian probability modelling to assign confidence intervals to each CRI score — distinguishing high-confidence findings from borderline signals that require additional data before clinical action. This engine is the reason we say no to treatment when others would say yes.

95% CI bandsPrior probabilityPosterior updateClinical action threshold
Engine 04
Active
Wearable Integration Matrix
WIM · Continuous Biological Surveillance

Blood panels are episodic. Biology is continuous. The WIM integrates HRV, sleep architecture, VO₂ proxy, glucose variability, and activity data from clinical-grade wearables — Apple Watch, Oura, Garmin, WHOOP — into the CRI score, filling the 23.5-hour gap between blood draws with longitudinal physiological signal.

HRVSleep stagesCGM proxyRespiratory rateActivity load

Built on defensible science,
not wellness consensus.

Population Analytics
100,000+ Individual Blood Panels
Reference ranges in standard medicine are built on relatively small cohorts. The CRI normative models were developed from a decade of real-world clinical data across diverse SEA patient populations, capturing age, metabolic phenotype, and lifestyle variables that academic datasets routinely exclude.
Biostatistics · Johns Hopkins
Rigorous Statistical Architecture
The scoring algorithms were built using formal biostatistical methodology — not intuition refined over years of practice, though that too. Population distributions, multivariate regression, and Bayesian updating are the mathematical substrate of every CRI score produced.
AI in Healthcare · MIT / MSK
Clinical Machine Learning
Pattern recognition across high-dimensional biomarker data is a machine intelligence problem. Training informed by Memorial Sloan Kettering and MIT Sloan AI curriculum ensures the feature selection and outcome weighting within each engine reflects oncology-grade analytical discipline.

What this system demands of us.

If we can't show it, we won't claim it
Every protocol recommendation generated by the CRI must be supported by a minimum confidence threshold in the BCC. If the data is insufficient or inconclusive, the system flags this — and we defer treatment until it isn't. Profit does not override probability.
If we can't help, we don't try
Patient qualification is gate-controlled by the CRI. Not every patient is a candidate for every protocol. We decline cases where our honest assessment is that the biological signal doesn't support the expected outcome — even when the patient wants treatment and has the means to pay for it.
Competitive pricing is a moral position
USP797-grade compounding and clinical-grade exosome protocols can be priced to exclude or priced to reach. We maintain the most competitive structure in SEA for this standard of care — because access to accountability-driven medicine should not require generational wealth.
First does not mean finished
Being the first to deploy a clinically scored regenerative index in Southeast Asia carries a specific obligation: to keep the evidence base current. The CRI updates as the science updates. The engines are living instruments, not published artefacts.
Outcomes over ROI.
All day long.
The operating principle of Dripdok · Est. Bali, Indonesia

The Close Regenerative Index exists because we believed — and still believe — that a patient's biological improvement is the only metric that matters. Every line of code in these engines, every biomarker weight, every Bayesian prior, was set in service of that belief. Not for a pitch deck. Not for a valuation. For the person sitting in front of us whose body is asking a question we are now, finally, equipped to answer with precision.

Ready to calculate your CRI score?

Enter your biomarkers and receive your personalised longevity score instantly.

Calculate My CRI Score →
We’ll reply when possible. Please follow instructions when prompted.
P
Open chat