AEGIS · RRAS Console

Nephrologist reviewing a renogram clearance curve and split-function chart on a diagnostic console

Catch the obstruction before the kidney fails.

Renogram washout kinetics plus a cheap lab panel — an obstruction call, a severity gauge, and resource planning across 24 countries, with a clinician in command of every decision.

1 · Patient — renogram + clinical snapshot

Enter the renogram kinetics and a cheap lab panel. The engine estimates function and detects obstruction — no nuclear scan needed for the severity estimate.

Renogram washout k

Split function (affected %)

Serum creatinine (mg/dL)

Age (yrs)

Blood urea (mg/dL)

Haemoglobin (g/dL)

Albuminuria grade (0–4)

Sex

MaleFemale

Pre-existing conditions

Diabetes mellitusHypertension

Result — obstruction call + severity

SVM decides obstruction; SVR grades kidney function (est. eGFR). Both defer at the margin.

Confidence in obstruction call (SVM)

Annual hospitalisation risk (reference)

Renal severity: est. eGFR (SVR)

eGFR from creatinine · KDIGO stage

severity index (100 = most severe)

⚠ Human-in-command.

2 · Medical resource planning — CKD / renal population

🌍 GLOBAL COVERAGE — planning anchored to the real health-system spend of 24 countries. Choose your market; renal cohort cost, bed-days, dialysis burden and avoidable spend re-compute in your own economy.

Renal / CKD-risk patients in your service

Market (per-capita cost anchor)

Annual care cost (renal cohort)—

Expected inpatient bed-days / yr—

Dialysis burden / yr—

Cost avoidable via early nephroprotection—

P1 · NASNurses for follow-up (@500/nurse)—

P2 · AdherenceNephroprotection reviews / yr—

Early-obstruction to pre-empt (shadow cohort)—

Ties the individual call to the system view: every early-obstruction kidney flagged here becomes nurse-follow-up load (Project 1, Nurse Augmenting System) and a nephroprotection-monitoring case (Project 2). Catching the rim before dialysis is where the avoidable cost lives.

How the numbers are derived — data vintage & cost method

Data sources & vintage

L1 · Renogram kinetics	Author's doctoral thesis — renogram simulations (Vol. 2, Appendix G). 40 fitted kidneys.
L2 · Screening labs	UCI Chronic Kidney Disease dataset (400 subjects), collected c. 2015.
L3 · Population risk	US CDC BRFSS survey, 2020 (Personal Key Indicators, 319,795 records).
L4 · Hospitalisation	UCI Diabetes 130-US Hospitals, renal-diagnosis subset (ICD-9 585/584/586/588/250.4/403–404). Encounters 1999–2008.
L5 · Economics	Country current health-expenditure per capita (WHO / World Bank), compiled mid-2026, converted to SGD.

How costs are derived — illustrative planning estimates, not billed figures:

Annual care cost	= patients × severity multiplier (Clear 1.0 / Early-obstruction 1.6 / Obstructed 3.5; cohort-weighted ≈ 1.32) × per-capita anchor.
Dialysis burden	= dialysis patients × SGD 45,000 / yr (in-centre haemodialysis anchor).
Avoidable cost	= early-obstruction patients × per-capita anchor × 40% progression reduction (SGLT2i / RAAS-blockade, literature).
Bed-days	= admitted patients × mean length-of-stay (from real 1999–2008 renal encounters) × (1 + 30-day readmission rate).

Cohort rates are measured from the RRAS-RP synthetic cohort (main.csv): early-obstruction 17.1%, obstructed 8.9%, admitted 7.0%, mean LOS 5.01 d, 30-day readmission 14.1%.

A nephrologist studying renogram waveforms on a diagnostic display, making the final decision

The system flags the obstruction. A clinician makes every call.RRAS surfaces its own uncertainty at the margin and hands the judgement to a human — augmenting the nephrologist, never replacing them.