1 of 12 finalistsClinical Decision Support
Kidney risk is hiding in data you already collect. NephroSense AI reads it.
Routine blood panels carry a kidney-risk signal the creatinine threshold cannot see. One platform surfaces it, for the clinic and for research.
Review queueIllustrative — synthetic data
Pt 04831routine visit
unremarkable · no flag raised
Pt 11274routine visit
unremarkable · no flag raised
Pt 02956routine visit
unremarkable · no flag raised
Pt 19408routine visit
→ eGFR + urine albumin (uACR)unremarkable · no flag raisedPrioritize — confirmatory kidney evaluation
Pt 07362routine visit
→ eGFR + urine albumin (uACR)unremarkable · no flag raisedPrioritize — confirmatory kidney evaluation
Pt 23517routine visit
unremarkable · no flag raised
Routine records. Nothing flagged.
- Evidence
- Retrospective feasibility demonstrated on a US adult clinical cohort. Independent external validation is in progress and is not yet complete.
- IP
- Patent-pending technology. Provisional patent filed March 2026.
Award · Peer-Review Presentations: Innovation Track
First place, Innovation Track
DCI Network AI Conference 2026 · Harvard Medical School, Boston ·
“Directing the Test That Was Never Ordered: Pre-Specified Temporal Evaluation of a Case-Finding Model for Albuminuria in Adults With Preserved eGFR”
Presented by Poornima Dubey, PhD, Co-Founder and Chief Executive Officer · one of three papers selected through double-blind peer review
Awarded at
Read the story →One Platform, Two Channels
One platform foundation. Separate models, workflows, and evidence paths.
Biopharma Research · near-term focus
STRATA
Open for sponsor collaborationCohort construction for kidney and cardio-renal programs
Estimate the candidate population a site actually holds before you select it, rather than after.
For Clinical operations · Feasibility · Clinical development · Medical affairs
Explore the research platform4 further programs, each carrying its stage
Health Systems · clinical foundation
VANTAGE
Available todayCKD risk triage from the panel you already drew
Risk triage at the point of care.
For Clinicians, primary care groups, FQHCs
How It Works
From the panel on file to a clinical next step.
- US adults with CKD
- 37M
- of them don't know they have it
- 87%
- 01
Use the panel already on file
Enter values or upload a lab report. No new tests, no new hardware.
- 02
Get a risk score in under ten seconds
With a confidence signal alongside, so the strength of the result is visible before you act.
- 03
Act on KDIGO-aligned next steps
Confirmatory testing and referral guidance, as a PDF or FHIR R4 summary, under physician review.
nephrosenseai.com
Clinical Considerations
High
Nephrology Referral
High-risk CDS flag plus abnormal routine labs warrant expedited physician review
Moderate
Blood Pressure Optimization
Target < 130/80 mmHg per KDIGO guidelines
Routine
Lifestyle Modifications
Dietary sodium restriction and regular physical activity
Illustrative — synthetic data. Not a real patient or a performance claim.
Latest
Where we've been lately
- 1 of 12 finalists
World Medical Innovation ForumBoston · 2026
Finalist · pitched live
Inaugural cohortIntellectual propertyPatent pending
Team
Built by scientists, physicians and AI engineers
Dr. Poornima Dubey, PhD
Co-Founder, Chief Executive Officer
MIT · HMS · IIT · AI/ML · Nanobiotech · HealthTech
Dr. Nisha, MBBS
Co-Founder, Chief Clinical Officer
KGMU · Clinical Oversight
AJ
Dr. Amit Jain
Co-Founder, Chief Technology Officer
CS/AI Engineer
Institutional Ties & Background
Backed by
Physician oversight required. NephroSense AI does not provide diagnoses and has not yet received FDA classification or clearance.
Advised by clinical, commercial and regulatory veterans. Meet the advisory board →
Ready to see what your routine data holds?
Start a conversation about a retrospective look-back, or try the clinical demo with built-in scenarios.