Oncology research,run by AI agents.
Insight on Oncology reads your papers and clinical data, designs the study, builds traceable cohorts, and returns publication-ready analyses, all in plain language.
No coding required · Literature to figures · Reproducible by default
Trusted by
Oncology research is stuck in a manual pipeline
From research question to manuscript for a typical retrospective oncology study.
Most of that time is non-scientific work:
- Harmonizing messy EHR data
- Hand-defining cohorts
- Mapping variables by hand
- Re-running analyses on every change
- Assembling reproducible docs for IRB, peer review & regulators
The bottleneck isn't ideas. It's the manual data-to-analysis pipeline.
Three core screens,one continuous flow.
Design the study, build a traceable cohort, and run reproducible analysis, each a dedicated screen, with you approving every step.
Analysis you can re-run
Models produce publication-ready figures with the code behind them. Open any chart, read the script, and reproduce the result.
Reproducible cohorts, by construction
Datasets and references become a versioned cohort with a CONSORT-style flow diagram and exported tables, with eligibility, grouping, and counts all traceable.
Protocol-grade study design
IO turns your question into a structured, SPIRIT-style protocol. Every section and item is drafted and editable, so the study plan is auditable from day one.
Frequently asked
ChatGPT and Claude are general-purpose assistants that answer in a chat bubble. IO is an operating system built for oncology research that runs on a workflow: query, cohort, analysis, validation, sign. Instead of generating prose, it executes real code on your clinical data and ships a pinned, version-locked, reproducible output at every step.
Most tools return only figures and p-values. IO ships the Python code, data mappings, execution log, and HITL approvals that produced the result. One Reproduce click rebuilds the same result anywhere.
Research data never leaves the site. Every analysis runs in an isolated sandbox, and only the code-set registry is shared across sites. Queries that risk PHI exposure are blocked before they run.
Ask your research question in plain language and IO decides the workflow and writes the code. But all code is fully exposed, so PIs fluent in statistics and programming can review, edit, and re-run it directly.
Data lineage, code versioning, HITL coverage, and external replication. These four axes are scored 0–100 and combined. Every output is signed and locked with content-addressed lineage.
The Spring 2026 pilot prioritizes PIs at NCI-designated and Korean academic cancer centers. Apply with a work email and we will reach out in order.