Capture the returned tray
One photo connects the meal served with what the patient actually consumed.
NUTRI-GUARD AI · AI FOOD INTAKE MONITORING FOR HOSPITALS
See malnutrition risk soonerBuild the evidence to act
Make intake visible. Keep dietitians in the lead. Nutri-Guard AI turns returned tray photos into measured intake data. Registered dietitians see declining intake sooner and keep complete documentation ready for clinical review.
SCAN. FIND. PROVE.Nutri-Guard AI workflow
01 · SCANCapture the returned tray
Surface nutrition risk sooner
Declining energy, protein, and food-item intake becomes visible for registered dietitian review.
Support the physician coding decision
Keep the clinical evidence and documentation available for physician review without asking dietitians to code.
SEE IT IN ACTION
From one tray photo
to clinical evidence
Watch the complete Nutri-Guard AI workflow in 15 seconds, from returned tray capture to documentation reviewed by clinicians.
Capture the returned tray
Link real meal evidence to the patient record.
Spot intake gaps
Review food item, energy, and protein intake.
Support physician coding review
Carry the documented clinical findings forward for physician review and coding.
FROM OBSERVATION TO EVIDENCE
The same tray becomes usable nutrition data
Nutri-Guard AI is an AI food intake monitoring system for hospitals.
It measures what inpatients actually eat from photos of returned trays and turns each meal into an intake record
for each patient that registered dietitians can review over time.
Today, intake is often a rough visual estimate (50%, 100%) written into the chart, not structured clinical data.
Food items and intake measured from photos become evidence for each patient that you can review across meals.
03 · CLINICAL REVIEW
Turn intake data into a review-ready decision
Longitudinal intake evidence gives registered dietitians a clearer basis to review risk and document their assessment.
FIND · Surface sustained intake gaps soonerRD REVIEW · Keep intake evidence with the assessment
THE GAP IN ONE VIEW
1 in 3 inpatients is at risk of malnutrition
Fewer than 1 in 10 is diagnosed and documented
An estimated 30% of hospitalized adults are at nutritional risk, while about 8.9% receive a coded malnutrition diagnosis.
That leaves a 21.1% gap between risk and documented care.¹
THE MALNUTRITION CARE GAP
30% AT RISK · 21.1% DOCUMENTATION GAP · 8.9% DIAGNOSED & CODED
21.1%DOCUMENTATION GAP
WHAT THE GAP MEANS
400-BED HOSPITAL · ILLUSTRATIVE
≈20,00001 · ANNUAL INPATIENT DISCHARGES≈4,00002 · PATIENTS AT RISK BUT UNDOCUMENTED≈$7.2M03 · FOR QUALITY & FINANCE LEADERS
Illustrative estimate: 400 beds × 50 annual inpatient discharges per bed × 21.1% documentation gap × $1,700 per case ≈ $7.2M annually. Varies by payer mix, MS-DRG assignment, and POA status.
Not a reimbursement guarantee. Sources: Journal of Hospital Medicine, 2024; Nutrition in Clinical Practice, 2021.
CMS MALNUTRITION CARE SCORE
Built for a more measurable malnutrition care process
CMS eCQM · 2026 REPORTING PERIOD
The Malnutrition Care Score (MCS, formerly the Global Malnutrition Composite Score) evaluates whether eligible adult encounters receive the appropriate sequence of malnutrition care.
Risk screening
Screen for malnutrition risk or place a dietitian referral.
Nutrition assessment
Confirm findings through an RD or RDN assessment.
Diagnosis
Document moderate or severe malnutrition when clinically appropriate.
Care plan
Document a current nutrition care plan for eligible patients.
Where Nutri-Guard AI fits
Nutri-Guard AI is an intake evidence and monitoring layer that supports nutrition assessment and care plan follow-up. It supports, never replaces, clinician judgment required documentation, EHR implementation, or CMS compliance review.
Review the official eCQM specifications ↗Built on the Nuvilab food AI platform
Nutri-Guard AI runs on Nuvilab's food AI platform, which has analyzed 100M+ meals across 1,300+ sites as of Sep 2026.
#11 globally · 2025
Innovation Award · 2024
Silver Award · 2023
Circular Economy cohort · 2023Built on Nuvilab's intake AI track record
Platform, research, and partnership references behind Nutri-Guard AI. Hospital validation specific to Nutri-Guard AI is in progress. Results will be published when available.
HOSPITAL · ACUTE CARE
AI intake validation
Intake measured in hospital settings is the direct foundation for the Nutri-Guard AI clinical workflow.
CLINICAL RESEARCH · ELDER CARE
Longitudinal intake research
Meal intake data supported a personalized nutrition intervention study across elder care facilities.
HOSPITAL · PARTNERSHIP
Workflow validation in progress
A hospital partnership exploring automated tray scanning as an alternative to manual estimation.
Need to verify trays before delivery? See Nutrition AI →Make intake visible
Make intake visible
Keep dietitians in the lead
Make every meal part of the care record
Contact us