ideasIdea
Every routine CT scan checked by AI for early cancer signs, with a tracked follow-up pathway
Hundreds of millions of CT scans are done each year for other reasons. Software could check each one for early lung, kidney, liver and pancreas changes, but only if a follow-up system exists.
Opportunistic screening algorithms exist for lung nodules, renal masses, liver lesions, and body composition. Propose a module running on all adult CTs with structured incidental-finding output, an automated tracking system to ensure follow-up, and a registry recording downstream procedures and cancers found so harm can be measured.
Hypothesis
Opportunistic AI on routine CT detects one to two additional stage I-II cancers per 1,000 scans, with follow-up completion of at least 90% when tracking is automated, versus under 50% today for incidental findings.
Rationale
Incidental findings are already common but get lost in free-text reports; the failure is tracking, not detection.
What would test it
Prospective pilot in two hospital systems (100,000 scans), measuring incremental cancers, stage, procedures per cancer, and lost-to-follow-up rate.
Maturity
early clinical
Who has to act
engineering
Cost to try
Medium ($1M to $50M)
Years to first evidence
3
Bottlenecks it attacks
- Most lethal cancers are found late · Screening exists for only a few cancers. Pancreatic, ovarian, liver, oesophageal and most lung cancers are found when cure is unlikely.
- Overdiagnosis and false alarms · Finding more cancer is not the same as saving lives. Screening also finds cancers that would never have hurt anyone, and treats them.