WHAT’S POSSIBLE
Your abstract, as a talk you could give tomorrow
A full slide plan, your opening and closing written out, and a four-minute emergency version.
Read & practise ↗Source: OpenAI · ChatGPT Health · F5GREYBRAIN FIELDNOTES
Your many roles.
One method underneath.
One stream shows you something you can make this week. The other looks at what is being claimed, and what the evidence actually supports. Both end at a skill you can practise.
WHAT’S POSSIBLE · A useful workflowWHAT’S ADVANCING · Evidence in context
THE EDITORIAL LIBRARY · RECENT FIELDNOTES
What changed. What you can do about it.
WHAT’S POSSIBLE
Rewrite it in Hindi. Then prove nothing changed.
A named list of exactly where your instruction lost its meaning — the step every translation tool skips.
Read & practise ↗Source: The Lancet Digital Health · F3WHAT’S POSSIBLE
Reviewer 2’s objections, before you submit
The three sets of objections you are going to receive anyway, ranked, while you can still do something about them.
Read & practise ↗Source: Nature Medicine · F1WHAT’S POSSIBLE
One case, turned into an hour of teaching
A timed session with the case staged for progressive disclosure, and the discussion mapped in advance.
Read & practise ↗Source: npj Digital Medicine · F2WHAT’S POSSIBLE
Your idea, in the shape a funder reads
Your idea reframed into aims and hypotheses, with the weak half named rather than flattered.
Read & practise ↗Source: JAMA Network · OpenEvidence · F3WHAT’S POSSIBLE
The photographs you keep meaning to take
A loose idea turned into twelve specific frames, a schedule, and a definition of finished.
Read & practise ↗Source: New England Journal of Medicine · F6WHAT’S POSSIBLE
The questions to take to your accountant
A structured question list and document checklist — so the hour with your CA is spent on answers, not on working out the questions.
Read & practise ↗Source: New England Journal of Medicine · F0WHAT’S POSSIBLE
A poster people stop at, designed in one shot
A layout, every block of text, the figure brief, and the forty seconds you say out loud.
Read & practise ↗Source: OpenAI · ChatGPT Health · F5WHAT’S POSSIBLE
Make it argue against your working diagnosis
A rehearsal of the argument against yourself, and the name of the trap you were walking into.
Read & practise ↗Source: The Lancet Digital Health · F3WHAT’S POSSIBLE
Audit your own methods like a journal would
Your methods section scored against the checklist a journal editor will actually open, before they open it.
Read & practise ↗Source: Nature Medicine · F1WHAT’S POSSIBLE
A teaching reel, with nothing identifiable in it
A shot-by-shot storyboard, three tested openings, and a de-identification pass over the model’s own script.
Read & practise ↗Source: JAMA Network · OpenEvidence · F4WHAT’S POSSIBLE
A complete OSCE station, built from nothing
Candidate sheet, actor script, weighted checklist and rating scale — the whole station, ready for the department to mark up.
Read & practise ↗Source: npj Digital Medicine · F2WHAT’S POSSIBLE
A national guideline, on one departmental page
A one-page protocol with every line traceable to the guideline — and an honest list of the places your setting does not match it.
Read & practise ↗Source: JAMA Network · OpenEvidence · F3WHAT’S POSSIBLE
The questions to ask before it is too late
An interview plan and forty specific questions — including the ten people most regret never asking.
Read & practise ↗Source: New England Journal of Medicine · F6WHAT’S POSSIBLE
Your clinic’s website, finished in one message
A finished single-file website you can open in a browser, with every blank listed for you to fill.
Read & practise ↗Source: New England Journal of Medicine · F0WHAT’S POSSIBLE
Your abstract, as a talk you could give tomorrow
A full slide plan, your opening and closing written out, and a four-minute emergency version.
Read & practise ↗Source: OpenAI · ChatGPT Health · F5WHAT’S POSSIBLE
Rewrite it in Hindi. Then prove nothing changed.
A named list of exactly where your instruction lost its meaning — the step every translation tool skips.
Read & practise ↗Source: The Lancet Digital Health · F3WHAT’S POSSIBLE
Reviewer 2’s objections, before you submit
The three sets of objections you are going to receive anyway, ranked, while you can still do something about them.
Read & practise ↗Source: Nature Medicine · F1WHAT’S POSSIBLE
Thirty days of posts, from what you already know
Thirty ideas cut down to the ten that fit your real week, with every risky claim flagged.
Read & practise ↗Source: JAMA Network · OpenEvidence · F4WHAT’S POSSIBLE
One case, turned into an hour of teaching
A timed session with the case staged for progressive disclosure, and the discussion mapped in advance.
Read & practise ↗Source: npj Digital Medicine · F2WHAT’S POSSIBLE
Your idea, in the shape a funder reads
Your idea reframed into aims and hypotheses, with the weak half named rather than flattered.
Read & practise ↗Source: JAMA Network · OpenEvidence · F3WHAT’S POSSIBLE
The photographs you keep meaning to take
A loose idea turned into twelve specific frames, a schedule, and a definition of finished.
Read & practise ↗Source: New England Journal of Medicine · F6WHAT’S POSSIBLE
The questions to take to your accountant
A structured question list and document checklist — so the hour with your CA is spent on answers, not on working out the questions.
Read & practise ↗Source: New England Journal of Medicine · F0WHAT’S POSSIBLE
A poster people stop at, designed in one shot
A layout, every block of text, the figure brief, and the forty seconds you say out loud.
Read & practise ↗Source: OpenAI · ChatGPT Health · F5WHAT’S POSSIBLE
Make it argue against your working diagnosis
A rehearsal of the argument against yourself, and the name of the trap you were walking into.
Read & practise ↗Source: The Lancet Digital Health · F3WHAT’S POSSIBLE
Audit your own methods like a journal would
Your methods section scored against the checklist a journal editor will actually open, before they open it.
Read & practise ↗Source: Nature Medicine · F1WHAT’S POSSIBLE
A teaching reel, with nothing identifiable in it
A shot-by-shot storyboard, three tested openings, and a de-identification pass over the model’s own script.
Read & practise ↗Source: JAMA Network · OpenEvidence · F4WHAT’S POSSIBLE
A complete OSCE station, built from nothing
Candidate sheet, actor script, weighted checklist and rating scale — the whole station, ready for the department to mark up.
Read & practise ↗Source: npj Digital Medicine · F2WHAT’S POSSIBLE
A national guideline, on one departmental page
A one-page protocol with every line traceable to the guideline — and an honest list of the places your setting does not match it.
Read & practise ↗Source: JAMA Network · OpenEvidence · F3THE EVERGREEN PRACTICE LIBRARY
Choose a role. Practise a skill.
Eight exercises that are always here, whichever hat you are wearing today.
EVERGREEN PRACTICE · THE ACADEMICIAN
Trace the claim before you trust the answer
There’s reading a paper. And then there’s really getting under its skin. Bring your curiosity. Learn a more deliberate way to explore.
The moment you recognise
A paper for tomorrow’s journal club. A dozen questions worth asking.
Try it with this synthetic material
A fictional journal-club abstract describes 48 residents whose mean quiz score rose from 6/10 to 8/10 immediately after a workshop. There is no control group, follow-up or patient-outcome measure.
Make something useful
Make a claim/evidence/limitation table. Identify why this cannot establish a lasting improvement in clinical care. Draft three questions for journal club.
Your check: Separate supplied facts from suggestions. Open the source, identify what is missing and record what you changed.
TWO MINUTES · NOTHING TO SIGN UP FORYour first win
Your first win
starts here.
An AI draft can overclaim confidently, in fluent English, with no warning that it has done so. This is the check that keeps the doctor in the loop.
No sign-up. You keep the template.
THE JOURNAL CLUB LABGuided demo · fictional case
- 01Read the source
- 02Spot the leap
- 03Keep the method
First, what does the source actually say?
FICTIONAL STUDY BRIEF
48 residents attended a journal-club workshop. Their average appraisal quiz score rose from 6/10 before the workshop to 8/10 immediately afterwards.
No comparison group. No later follow-up. Patient outcomes were not measured.
Now imagine handing this to an AI and asking for a summary for tomorrow’s journal club. Your job is what comes next: checking what it hands back.
This example is prewritten. No live AI request or real study is involved.
Turn a useful habit
into something you can show.
Self-paced foundations. Private capstone work. An exam and assessed progression towards the IIHMR Bangalore certificate programme. Eligibility and programme terms apply.
Explore the learning and certificate pathway ↗