How NephroSense AI Works
A simple, three-step workflow that turns routine lab data into clinician-ready CKD decision support.
We don't replace any tests. We connect existing data faster, so the right patients get follow-up sooner.
Input
Routine record
Entered directly, uploaded as a lab report, or a built-in demo scenario.
Assessment
Risk score + confidence signal
Returned with a confidence interval and an uncertainty level.
Output
Considerations · PDF & FHIR R4
KDIGO-aligned clinical considerations, ordered by urgency, for physician review.
The Platform
One platform, two workflows
The same platform serves the clinical workflow below and a research workflow for retrospective sponsor studies — separate models and separate evidence paths per application, sharing the routine-panel inputs, the deployment architecture and the IP.
- Inputs
- Values already present in a routine adult workup — standard blood work, vitals and basic history, drawn and recorded as part of ordinary care.
- Integration
- Output is built on HL7 FHIR R4 for clinical data exchange. Embedded EHR deployment is in development.
- IP
- Patent-pending technology. Provisional patent filed March 2026.
Clinical Workflow
The clinical workflow (VANTAGE)
Three steps. The walkthrough shows what the clinician actually receives at the end of them.
Enter Patient Data
Choose how you provide biomarkers: type them in manually, upload a lab report PDF or image for automatic extraction, or use one of three built-in adult demo scenarios.
Receive Risk Assessment
Results come with a confidence signal and help clinicians prioritize which patients to follow up with first.
Act on Clinical Guidance
Use the flagged result to prioritize follow-up, closer monitoring, and physician-reviewed care, with a PDF report or FHIR R4 export for the EHR.
See NephroSense AI in Action
A guided walkthrough of the routine-lab workflow, on synthetic cases. This is the walkthrough this page promises.