Resident · MSc ResearchThesis / dissertation
First synopsis, alone, without methodological support.
- Framing the research question (PICO)
- Sample size justified with references
- Synopsis ready to submit to the thesis director
Hosted in FranceGDPR compliant · MR-004· Methodological platform for clinical research
Your synopsis, ready to submit before the end of the week.
Protocolis is the methodological assistant for hospital investigators. The AI asks the right questions, structures the synopsis according to your regulatory framework, and delivers a compliant Word document — hosted in France, your content never reused for training.
CoversRIPH 1RIPH 2RIPH 3RNIPHDrug studiesMD / IVD-MD
No credit cardCancel in 1 clickIsolated per institution
protocolis.fr · studies · REHAB-COLO-2026
Illustrative previewHosted in FranceAssistant online
A prospective single-center trial on rehabilitation after colorectal surgery. For the primary endpoint, which primary indicator are you considering?
Length of stayClavien-Dindo complications ≥ IIReturn of bowel function
Synopsis · v3
3 / 12 sections
Study title
Scientific background
Primary & secondary objectives
Primary endpoint
Population & inclusion criteria
Sample size calculation
Statistical plan
Ethical considerations
Built with hospital research-office methodologistsCo-development underway with several pilot centers — references publishable once consent is formalized.
The regulatory framework
You describe your research. Protocolis infers the regulatory framework.
As you describe your project (population, data collected, patient pathway), the assistant identifies the applicable research category and the authorization circuit to follow.
Categories Protocolis can recognize
RIPH 1
Interventional research
Change to patient care, non-minimal risk. Ethics committee + ANSM circuit.
RIPH 2
Minimal risks & constraints
Acts outside routine care with low risk. Ethics committee only.
RIPH 3
Non-interventional
No change to patient care — observations, questionnaires, negligible sampling.
RNIPH
Research not involving the human person
Reuse of already-collected data, retrospective studies. Reference methodology framework (MR-004).
Drug studies
Research on medicinal products
Specific regulatory framework — RIPH 1 with ANSM requirements, investigator brochure, pharmacovigilance.
MD / IVD-MD
Medical devices & IVDs
MD clinical investigation or IVD-MD performance study. MDR / IVDR regulation + ANSM.
Is your case unusual? The team will look into it with you.
Method
Three steps. No blank page.
01
Describe your project
A few sentences: your research question, your study population, your endpoint. No form to fill in blindly.
02
Build through conversation
The assistant follows up section by section, proposes methodological wording, searches PubMed and ClinicalTrials, computes the required sample size.
03
Export or share
Download the Word document compliant with research-office standards, or send it directly to your methodologist or department head via a secure link.
What Protocolis brings you
A methodologist who never sleeps. Not a chatbot.
The assistant probes the grey areas.
Rather than a blank page, Protocolis asks the questions a methodologist would ask — in order, with examples tailored to your specialty.
- Automatic detection of missing sections
- Suggestions contextualized by specialty
- Sample size and proposed statistical hypotheses
- Timestamped conversation history, exportable
You described an open-label design. For a comparative trial, a control arm is expected. Do you prefer individual or cluster randomization?
Individual randomization, 1:1.
Perfect. With a binary primary endpoint and an assumed effect of 15%, I propose n = 212 (two-sided α = 0.05, power 0.80). Shall we validate?
Illustrative preview. Protocolis outputs constitute a methodological working basis — final validation remains the responsibility of the reference methodologist.
Security · Sovereignty
Frame AI within your institution.
Your investigators already use ChatGPT to write, often outside any framework. Protocolis turns this unmanaged, off-the-books use of AI into a compliant, auditable channel hosted in France.
Hosting in France
Your data is hosted in France and encrypted in transit (TLS). Calls to our AI subprocessors are governed by standard contractual clauses.
Per-user data isolation
Each study is isolated to its author; no exchange between users without explicit, revocable sharing. Your protocols stay confidential from teams competing on the same calls for proposals.
Your content is never used to train AI
Contractual guarantee. Independent annual audit.
Research-office audit logs
Who generated what, when. Configurable retention.
TLS encryptionGDPRDaily backupsFR sovereigntyDPA available
Request an institutional walkthrough Institutional offer available
The team
MD
Marie De Solère
Promotion coordinator · DRCI, GHICL · 9 years supporting PHRC-N
« I have reviewed hundreds of synopses. Protocolis does what I used to do by hand — without the queue. »More about the team
Representative use cases
From resident to professor, adaptive methodological and regulatory support.
Three usage patterns observed with our pilot investigators.
Junior hospital practitionerGrant call · tight deadline
First grant-call submission, without a dedicated methodology team.
- Research question, criteria and design framed in one session
- Sample size and analysis plan justified with references
- RIPH regulatory classification proposed and documented
Professor / senior practitionerPHRC-N grant assembly
Upstream framing before the biostat / methodology team, under grant-call timing.
- Structuring the synopsis before the team meeting
- A framing baseline for supervised juniors
- Keeping methodological iterations
Frequently asked questions
Everything investigators ask us.
Protocolis processes no patient data. The tool produces methodological content (synopsis, protocol, bibliography) from your description of the project — not from individual data. Furthermore, your content (conversations, synopsis, imported documents) is never used to retrain the models. Contractual guarantee, isolation per institution workspace, hosting in France.
Protocolis does not replace your methodologist — it saves you the hours of upstream structuring, so the methodological review focuses on substance. The CRAs at pilot centers use it as a reference draft.
Protocolis proposes the sample size, the hypothesis, the test and the justification — based on the cited literature. For complex designs (cluster, adaptive sequential, Bayesian), it asks the right questions and structures your later exchange with your biostatistician.
Each study is isolated to your account (or to your institution's workspace on an institutional license). No leakage between institutions, no external access — including for our team, except with explicit consent for support.
You resume the conversation with the assistant by submitting the research-office feedback — it generates a revised version point by point. All iterations are kept and exportable.
Word (.docx) in French, with layout compliant with research-office standards. PDF, English version, PHRC-N letter of intent and ethics-committee form are being added.
Another question? Write to us — answer within 24 business hours.
Ready to start
Your next synopsis, this afternoon.
7 days free, no credit card. Invite your team or your research office when you are ready.
No credit card · Cancel in 1 click · Hosted in France