Raison d'être
Five billion people cannot access safe surgical care. The bottleneck is the training pipeline, and simulation is the lever.
Five billion people cannot access safe, affordable surgical care.
The number comes from the Lancet Commission on Global Surgery. It does not describe people waiting a few weeks for an elective procedure. It describes people for whom appendicitis, an obstructed labour, a hernia, or a fractured femur is an emergency without a resolution. The Commission estimates that roughly 17 million people die every year from surgically treatable conditions, more than the combined toll of HIV, malaria, and tuberculosis.
The burden is not distributed evenly. A third of the world's population lives in low and middle income countries, and only about 6 percent of the world's operations happen there. Surgeon density in the poorest countries is under 0.2 per 100,000 people; in high income countries it is often more than ten times that.
To close the gap, the Lancet Commission set a target of at least 20 surgeons, anaesthetists, and obstetricians per 100,000 people by 2030. Many low income countries today have fewer than one. Meeting that target requires roughly doubling the global surgical workforce over the coming decade. Nothing in the current training pipeline is going to produce that.
The bottleneck is the pipeline, not the will
The reason we do not have more surgeons is not that no one wants to train them. It is that training a single surgeon is one of the longest and most expensive human capital investments in medicine.
In the United States, a general surgeon spends 13 to 15 years in training: four years of undergraduate study, four of medical school, five of residency, and often another year or two of fellowship. The direct educational cost, before residency salary, runs between about $290,000 and $550,000 in tuition and fees. In the UK the pathway looks different but the total is comparable in years, and the marginal cost to a health system training a consultant surgeon is on the same order of magnitude.
The training is long because so much of it depends on real patients. A trainee acquires competence one operation at a time, under the supervision of an already qualified surgeon, on a case list that arrives at the speed of biology. There is no equivalent, in surgery, of the flight simulator that took commercial aviation from an apprenticeship into an industrialised profession.
That is the bottleneck. Not classroom knowledge, not motivation. The fact that every hour of meaningful surgical practice, for most of a decade, is gated by the availability of a mentor, a theatre, and a case.
Simulation compresses the curve
There is now a substantial evidence base that simulation shortens the surgical learning curve without sacrificing outcomes. In a prospective study of laparoscopic cholecystectomy, trainees who completed a proficiency based simulation curriculum before operating on patients had 4.5 times fewer adverse events than a control group taught by conventional apprenticeship, and they started clinical training at a higher baseline of skill. Systematic reviews across laparoscopic hernia repair and other minimally invasive procedures report the same pattern: fewer complications, shorter operating times, faster time to independent practice.
The simulators used in those studies are, by the standards of modern computing, very limited. They are box trainers and rigid virtual reality kits with narrow behaviour and no intelligence of their own. If that evidence base already exists for the current generation of tools, the question is what becomes possible when a simulator can watch, evaluate, and adapt.
What OpenSurgery is building
OpenSurgery is a research lab. The thesis is straightforward.
If we can build simulators that are realistic enough to substitute for a meaningful fraction of in theatre training, and intelligent enough to teach, then the cost of producing a surgeon falls and the time to competence shortens. The training pipeline widens. More surgeons are trained per year, more health systems can afford to train them, and more of the five billion get access to safe surgical care.
The first project is a laparoscopic cholecystectomy simulator. Cholecystectomy is the most common laparoscopic operation in the world, it is safety critical, and the learning curve is already well characterised in the surgical literature. It is the right place to start because progress is measurable and the population who benefit is large.
The longer bet is foundation models for surgery, and that is the subject of future writing here. For now, if you would like to hear when the simulator is ready, or when we publish the next piece of research, the mailing list is on the homepage.