Open access. Appraised, not advertised.Practise the cut before you make it.
Every simulation model we can verify, in one place, with its real cost, its materials and an honest reading of the evidence behind it.
Open access, and no account needed. All we ask is that you pass it on to your colleagues. No model is listed here because a manufacturer asked for it.
Simulation research is done, published, and then lost.
Somebody in Addis Ababa builds a bowel anastomosis trainer out of abattoir tissue for under ten dollars, runs a course with it, and writes it up. Somebody in Leiden pits a portable robotic simulator against a console costing six figures and finds it holds its own. Somebody in Chennai works out which of four cheap materials actually behaves like gingiva. All of it is real, all of it is useful, and almost none of it is where a trainee would ever find it.
It sits behind paywalls, scattered across specialty journals that nobody outside that specialty reads. A dental student looking for a suturing model has no reason to search the maxillofacial literature. A surgical trainee in a hospital without a skills lab has no way to know that the thing they need was published, validated and forgotten eight years ago.
SimulationM8 collects it. Organised by specialty, tagged by what it actually costs, and appraised honestly so you can tell the difference between a model somebody tested properly and a model somebody simply liked the look of.
We say how good the evidence is, including when it is thin
Most published simulator studies still use validity language the field abandoned years ago. One review found that only 6.6 per cent used the current framework. So every appraised model here carries three separate readings rather than a single verdict.
The closest CASP checklist
Applied where it genuinely fits, and we say so plainly where it does not. Most of this literature is a before and after study or a build note, not a randomised trial.
A level of evidence
Oxford CEBM style, from Level 5 for a build method with no outcome data, through to Level 1 for a proper systematic review.
A Kirkpatrick level
Because a study can be methodologically sound and still only measure whether trainees enjoyed it, rather than whether anything changed.
A model with one satisfaction survey behind it is labelled as having one satisfaction survey behind it. We would rather under claim than over claim.
Frugal where it counts, honest where it does not
Most simulation is genuinely expensive. Commercial trainers, virtual reality platforms and cadaveric labs all have their place, and pretending otherwise would make this library less useful, not more principled.
So everything is catalogued regardless of price, and the models that are genuinely frugal, buildable from materials most trainees could source in an afternoon, carry a highlighted tag. Filter by it if that is what decides things for you, ignore it if it is not.
Robotic surgery is where this matters most. Almost all the evidence sits behind premium hardware, and yet a randomised trial recently put a portable, affordable simulator against a market leading console and it held its own.
Sixteen specialties, and dentistry is not a footnote
Generic surgical skills sit alongside dentistry, oral and maxillofacial surgery, and every subspecialty we have found published evidence for.
Generic surgical skills
FoundationsDentistry
FoundationsEmergency & critical skills
SpecialtiesENT & head and neck
SpecialtiesLaparoscopic & MIS
SpecialtiesLow resource curricula
SpecialtiesNeurosurgery
SpecialtiesObstetrics & gynaecology
SpecialtiesOphthalmology
SpecialtiesOral & maxillofacial surgery
SpecialtiesOrthopaedics & spine
SpecialtiesPaediatric surgery
SpecialtiesPlastics & flaps
SpecialtiesRobotic surgery
SpecialtiesUrology
SpecialtiesVascular access
Access to surgery is a training problem before it is anything else
To collect every simulation model the published literature can support, appraise it honestly, and put it within reach of any surgical or dental trainee, whatever their institution can afford.
The Lancet Commission on Global Surgery put it at five billion people without access to safe, timely, affordable surgical and anaesthesia care, and set a benchmark of twenty surgeons, anaesthetists and obstetricians per hundred thousand people. Across low and middle income countries the real figure sits at around three.
That gap is a workforce gap, which makes it a training gap. And there is evidence that this is not abstract: a study of minimally invasive surgery outcomes found that the presence of local simulation training facilities was independently associated with fewer complications, with the greatest benefit in hospitals with the least infrastructure.
Simulation is not a training luxury in that context. It is patient safety infrastructure that most of the world cannot currently afford, which is a good reason to stop keeping the recipes behind paywalls.
Before you use anything here
What does it cost to use the library?
Nothing. The appraised library is open access and requires no account. The only thing we ask in return is that you share it with your colleagues and peers, because the trainees who need it most are the ones least likely to stumble across it.
Who decides what gets listed?
Anything published that we can verify and appraise. No manufacturer pays to appear, and commercial systems are appraised on the same terms as a model built from an orange. Where the evidence for an expensive platform is genuinely strong, we say so.
Did you read every paper in full?
No, and each card says which. Screening and most appraisals are made from the abstract, which is enough to decide whether a model belongs here and to describe what was measured. Where that is the case the card says so and lists it as a limitation. Entries appraised from the full text are marked as such.
What does the frugal tag actually mean?
Buildable from materials most trainees could source in an afternoon, at near zero, under twenty dollars, or free and open source. It is a filter, not a judgement. Some frugal models have better evidence behind them than expensive ones, and some do not.
Can I submit a model I built?
Yes, and submissions from settings the published literature under represents are the most valuable thing we could receive. You need a materials list, a rough cost, a build method and whatever evidence exists, even if that is only informal trainee feedback. We will label it honestly either way.
Is this clinical advice?
No. It is educational material for students and clinicians. Nothing here is advice for the care of an individual patient, and no model listed here is a medical device.
Find the model, build it, practise the thing.
40 models, sixteen specialties, appraised and free.
Browse the library