ideasIdea

A consented commons of surgical video linked to pathology and outcomes

Record cancer operations (with consent), link each video to the pathology report and the patient's recovery, and open the collection to researchers to learn what surgical technique actually works.

Robotic and laparoscopic cancer surgery is routinely video-recorded, yet the recordings are deleted. A governed commons linking de-identified surgical video to margin status, lymph node yield, complications and survival would allow surgical technique to be studied at scale and surgical AI (phase recognition, skill assessment) to be validated against outcomes. Academic skill-assessment studies have already shown that surgeon skill rated from video predicts outcomes.

Hypothesis

Video-derived measures of technique and skill will explain a meaningful share of between-surgeon variation in positive margins and complications, and feedback based on them will reduce that variation.

Rationale

Birkmeyer et al. (NEJM 2013) showed peer-rated surgical skill from video predicts complication rates; the tools to scale this with AI exist but lack data.

What would test it

Collect 2,000 consented robotic prostatectomy or colectomy videos across five centres with linked outcomes; test whether automated skill scores predict positive margins and 30-day complications.

Maturity

speculative

Who has to act

research

Cost to try

Medium ($1M to $50M)

Years to first evidence

Bottlenecks it attacks

Data silos · Records, scans, genomes and outcomes sit in separate systems that cannot talk. Every patient's experience is lost to the next.
Surgery and radiotherapy cure most, get least · Surgery and radiotherapy cure more people than drugs do, but attract a fraction of the research investment.

Connected

6top

Pages like this

not linked directly; found by shared links

A consented commons of surgical video linked to pathology and outcomes

Connected

Pages like this

fronts

technologies

companies

bottlenecks