Medical-AI · SaMD · Clinical-Evidence

I design, run, and sign as the named methodologist the clinical-validation study at the core of your AI / SaMD regulatory dossier.

Biomedical engineer (BME MSc); has built and peer-reviewed a medical-AI device. Standalone validation study, CER statistical core, and real-world evidence for TITCK-CDSS, EU-MDR, and FDA dossiers.

The regulatory firm sets up the process, the AI engineer builds the model. I build the statistical evidence layer in between.

Built a Device

A medical-AI device was developed and published in an international peer-reviewed journal; the method and regulatory discipline transfer independently of the modality.

Engineer + Methodologist

The profile MEDDEV 2.7/1 Rev 4 looks for in a named clinical evaluator: relevant biomedical / biostatistics competence and proven methodology experience.

ML-Validation Literate

TRIPOD-AI, PROBAST-AI, DECIDE-AI; standalone performance, MRMC reader study, bias and representativeness methodology.

Regulation-Aware

Evidence framed to the statistical requirements of TITCK-CDSS (Jan 2025), EU-MDR, and FDA AI/ML guidance.

Services

A featured service, a solid core.

Each service is delivered as a fixed-scope package. Scope and budget are clarified in the free scoping call.

Featured · C1

AI / SaMD Clinical-Validation Study

3-8 weeks

I design the standalone clinical-validation study at the core of your artificial intelligence medical device (SaMD) regulatory file, run its statistics, and sign as the named methodologist (the statistical-methodological core; alongside your clinical evaluator when the file requires it). For the core of TITCK-CDSS, EU-MDR, and FDA files.

  • ·Study design + statistical analysis plan (SAP)
  • ·Standalone performance: AUC-ROC, calibration, decision-curve, MRMC reader study
  • ·Sample size / power, subgroup, and bias / representativeness analysis (TRIPOD-AI / FDA)
  • ·Independent test-split and external-validation cohort design

Budget

Scoped on a call

Request a quote →

C2

Clinical Evaluation Report: Statistical Core

4-10 weeks

I write the clinical-data and statistical core of the CER: systematic literature appraisal, clinical-data analysis, and the performance-evidence section. Not the QMS / technical-file stack, but the scientific core of the evidence.

  • ·Systematic appraisal of clinical evidence (MEDDEV 2.7/1 Rev 4)
  • ·Clinical-data / performance statistics section
  • ·Equivalence and literature-route evidence analysis
  • ·Statistical plan input for PMS / PMCF

Budget

Scoped on a call

Request a quote →

C3

Real-World Evidence (RWE) Protocol + SAP

3-8 weeks

Design, protocol, and analysis plan of a real-world evidence study for regulation and publication. Not data collection / execution; the design and statistics layer.

  • ·RWE study design + endpoint selection
  • ·Statistical analysis plan (SAP), confounding / bias strategy
  • ·Retrospective cohort / register design
  • ·FDA RWE (Dec 2025) and MDR-compliant framing

Budget

Scoped on a call

Request a quote →

C4

CRO Biostatistics: Subcontract

Flexible

I provide CRO and sponsor teams with the artificial-intelligence-validation and biostatistics depth your trial team is missing. Not a competitor, but an expert layer sold into the team.

  • ·SAP, sample size, randomization, endpoint methodology
  • ·Statistical report / TLF design consulting
  • ·Method depth in AI / diagnostic-accuracy studies
  • ·Named biostatistician (not PI / coordinator)

Budget

Scoped on a call

Request a quote →

Entry Point

Validation-Readiness Review

A fast, fixed-scope scan of the evidence, data, and method gaps in your AI / SaMD product. Before you enter the file, you get a clear view of where you stand and what is missing.

Fixed-scope intro package

Monthly Retainer

Fractional Head of Clinical Evidence

For seed / early-stage AI-health companies that cannot hire a full-time clinical-evidence / biostatistics expert, monthly, comprehensive on-call method and evidence consulting. About 1 day per week.

Monthly retainer

Out of Scope

What I do not do.

I deliberately work in a narrow area of expertise. I do not do the following; I refer them to suitable firms.

  • ×ISO 13485 / QMS quality-system setup, audit, and certification
  • ×TITCK / UTS product registration / licensing paperwork
  • ×CRO operations, site monitoring, trial execution
  • ×Clinical research principal investigator / coordinator / consent (requires a physician)
  • ×Therapeutic areas where I develop my own product (conflict of interest)

Frequently Asked

Frequently asked questions.

Who do you run the AI / SaMD clinical-validation study for?

For artificial intelligence medical device (SaMD) and medtech companies facing a regulatory dossier (TITCK-CDSS, EU-MDR, or FDA). The typical client is an FX-revenue or investor-backed seed / early-stage AI-health company making its first submission, without an in-house biostatistics / clinical-evidence expert.

You are not a physician; can you carry out this work?

Yes. My role is the named methodologist/biostatistician who designs and runs the study's statistical-validation evidence; I am not the principal investigator, coordinator, or consent-taker of the clinical study (those roles require a physician). My signature covers the statistical-methodological core and, when the dossier requires it, is presented together with your named clinical evaluator/physician. MEDDEV 2.7/1 Rev 4 defines the named clinical evaluator through relevant higher education and demonstrated experience. I have built and peer-reviewed a medical-AI device; I construct validation evidence through the eyes of both an engineer and a methodologist.

Do you set up ISO 13485 / a quality management system (QMS)?

No. ISO 13485 / QMS setup, auditing, and certification are outside my scope; I refer this work to suitable firms. My quality domain is not the factory but the methodological quality of the evidence and the model: ML-validation rigor, data / ground-truth / bias, reproducibility, and statistical integrity.

How does pricing work?

Every dossier is different; fees are given per service as a fixed-scope package after a free scoping call (not hourly). We clarify the scope and budget together during the call; payment is set in USD and can be invoiced in TL at the TCMB exchange rate.

Is there a conflict of interest if there is an area where you develop your own product?

In the therapeutic areas where I develop my own product, I do not provide consulting, to avoid a conflict of interest. In all other SaMD / AI-health areas I work, and every engagement runs under confidentiality (NDA).

First Step

Let us talk for 15 minutes first.

We clarify what stage your dossier is at and what is missing, and produce an impartial roadmap. All work runs under an NDA and a KVKK/GDPR data-processing agreement.

About the practitioner →