Zenlo Open Labs

Open science

Zenlo Research — autonomous AI pipeline & open publications

An autonomous research agent that scans literature, tests hypotheses on population data, and drafts journal-aware manuscripts — with human review at every stage. Methods and preprints are published in the open.

AI models5
NHANES adults4,018
Best-model F10.963
Biomarkers (engine)102
Research agent · three stages Literature → Analysis → Writing
1Literature AgentPubMed gap scan → structured hypotheses.
2Analysis AgentNHANES / BRFSS statistics → tables & figures.
3Writing AgentJournal-aware draft → human edit & submit.

How the research agent works

Three agents, one pipeline

Each stage produces artifacts the next stage can audit. The pipeline accelerates literature review and analysis — it does not replace scientific judgment.

Stage 01

Literature Agent

Scans PubMed and related sources for gaps, contradictions, and under-tested associations. Outputs structured hypotheses with citations — ready for statistical testing, not for automatic publication.

PubMed gap analysis · adversarial hypotheses · citation graph
Stage 02

Analysis Agent

Runs reproducible statistics on NHANES and BRFSS cohorts: logistic regression, trend analysis, chi-square tests, and ROC curves where appropriate. Results are logged for human review before any claim enters a draft.

NHANES · BRFSS · logistic · trend · chi-square · ROC
Stage 03

Writing Agent

Produces a journal-aware manuscript draft with real citations pulled from the literature stage. Tone, section structure, and disclosure language follow the target venue — then a human editor takes over.

Journal templates · citation integrity · disclosure blocks

Human-in-the-loop. Zenlo Research proposes analyses and drafts; scientists and clinicians decide what is sound, what to run next, and what gets submitted. We do not present agent output as peer-reviewed fact.

15 clinical patterns. 8 benchmarked across 5 AI models on 4,018 NHANES adults (F1 up to 0.963). Per-pattern NHANES audits published for 2 (Insulin Resistance, Metabolic Syndrome); the rest are structural CHAI cards, validation in progress.

Publications

Evidence you can verify yourself

Every major claim links to a public source. No gated PDFs — open science for open decisions.

Preprint · medRxiv · Feb 2026 · CC-BY 4.0Multi-Model Clinical Validation of an AI-Powered Biomarker Analysis Framework: A Cross-Vendor Benchmark on 4,018 NHANES PatientsDOI 10.64898/2026.02.13.26346284Five frontier models benchmarked on a real NHANES cohort — methodology, F1 comparison, and limitations published for independent review.Peer review · JAMIA OpenMulti-model validation of automated biomarker pattern detectionManuscript under review · Oxford University PressSubmitted to JAMIA Open — we will add a public link when a DOI is available.Article · Medium · Feb 2026We Tested 5 AI Models on 4,018 Real PatientsPublic methodology · readable summaryPlain-language walkthrough of the benchmark design and what we learned running five vendors on the same patient cohort.

Read the science behind Zenlo.

Start with the medRxiv preprint for the full methods, or the Medium article for a shorter overview. Clinical pattern audits for Labs live at /labs/audits.

Read the preprint → Read on Medium

→ medRxiv (CC-BY 4.0)→ JAMIA Open · under peer review→ Medium article