Evidence

Case studies built from inspectable evidence

I publish a small number of case studies in detail rather than a long list in summary. Each one shows what happened, what the evidence supports, and what remains uncertain. This page will grow slowly and deliberately.

What you will find here

The full working method, not a summary

How the system actually ran, including the parts that were simplified away and the parts that were deliberately kept.

Real corrections, including the rejected ones

Where clinical review changed the output, with the superseded versions preserved rather than quietly discarded.

Stated limitations

What the retained evidence does not establish, named explicitly rather than left out.

Featured Case Study

Medical AI governance · Gallbladder Surgeon Abu Dhabi

From Complex Automation to Clinician-Governed Medical Content

An AI-assisted medical content workflow was simplified from a multi-stage orchestration into a single principal drafting step — without removing the source evidence, structured claims, deterministic checks, clinical review, bounded correction or explicit publication authorization.

  • 16 English articles at the verified 17 August 2026 repository snapshot
  • Three retained correction cycles, rejections included
  • One five-gate approval record
Read the case study

Further case studies will be added as work reaches a point where the evidence can be published responsibly.

Get In Touch

Working on something where clinical judgement matters?

Tell me what you are building and where a practising clinician would be useful.

Discuss medical AI evaluation work