OnCo
ideasIdea

Validate and reimburse AI contouring and planning to expand radiotherapy capacity

Drawing targets and planning radiotherapy takes hours of scarce expert time. Properly tested AI could do much of it, letting the same staff treat far more patients, if regulators and payers set clear rules for proving and paying for it.

A programme with three parts: prospective, multi-centre validation studies of auto-contouring and auto-planning against expert consensus with clinical-acceptability and time-saved endpoints; a regulatory pathway that accepts these endpoints for clearance; and payer recognition that shifts payment from planning time to plan quality, so that time saved translates into more patients treated rather than lost revenue. Radiotherapy workforce shortages (physicists, dosimetrists, radiation oncologists) limit capacity in rich and poor countries alike, and AI planning is the most mature clinical AI in oncology, yet its adoption is slowed by unclear evidence standards and payment rules.

Hypothesis
Validated AI contouring and planning reduces planning time per patient by more than half without loss of plan quality in prospective multi-centre studies, and departments adopting it under supportive payment rules increase patients treated per staff member by at least a fifth within two years.
Rationale
Randomised and prospective studies of auto-contouring already show large time savings with acceptable quality for several sites; the barrier is systemic. Radiotherapy is the one area where AI could directly relieve a workforce constraint that rations curative treatment.
What would test it
Run a prospective multi-centre validation with time and quality endpoints, then a stepped-wedge implementation across ten departments measuring throughput and staff time.
Maturity
being tested at scale
Who has to act
regulator
Cost to try
Medium ($1M to $50M)
Years to first evidence
3
Bottlenecks it attacks

Connected

10top