NUTRIGRAPH · OS
ICMR × Gates Foundation Grand Challenge

An AI, Quantum & Agentic ecosystem to end anaemia in adolescent girls and WRA.

NutriGraph OS turns aspirational, iron-dense food into measurable haemoglobin gain — with knowledge graphs, RAG, agentic supply, digital twins, and quantum bioavailability simulations.

~50%
Anaemia prevalence in adolescent girls & WRA
4–5 mg
Bioavailable iron per serving target
12 wk
Digital-twin haemoglobin horizon

The stack, end-to-end

Every RFP criterion is mapped to a working module in the console.

AI Insights

Segment WRA by dietary context and psychological barriers.

Quantum Model

In-silico phytate ↔ iron docking before wet-lab trials.

Knowledge Graph

Hyper-local map of foods, nutrients, and barriers.

AI Assist (SLM)

Stigma-free chat over low-bandwidth channels.

Inferencing Nudge

Reframe consumption from selfish to essential.

Digital Twin

Per-persona haemoglobin projection over 12 weeks.

Agentic AI Supply

Autonomous stock rerouting across Anganwadi + retail.

Vector DB + RAG

Retrieval over NFHS + behavioural corpus.

Neural Feedback Loop

RL rewards adapt nudge strategy per persona.

Data Pipelines

NFHS → segments → twins → nudges.

Neural Network OS

Real-time telemetry for reviewers.

RFP → NutriGraph

A direct mapping from the ICMR × Gates evaluation criteria to what the console shows.

Target & gaps
Knowledge Graph + hyper-local nutrient gap heatmap
Innovation
GAN-style taste masking + Inferencing nudges over WhatsApp SLM
Scientific rationale
Quantum bioavailability model with rationale generator
Feasibility & scale
Agentic AI supply autonomously reroutes stock
M&E
Neural Network OS with real-time adherence telemetry
Team
Quantum chemists, behavioural DS, AI ethicists