# AI-assisted central imaging reads to cut endpoint cost and variability

Source: https://onco.cc/ideas/idea-tr1-ai-central-imaging-reads/  
OnCo record `idea-tr1-ai-central-imaging-reads` (Idea). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Measuring tumours on scans for trials is slow, expensive and inconsistent between readers. Software that measures lesions and flags changes, checked by a radiologist, could make trial endpoints cheaper and more reliable.

## Summary

Validated AI segmentation and lesion-tracking tools perform RECIST 1.1 measurements with radiologist adjudication only on flagged discordances, replacing dual blinded independent central review. Performance is locked and validated against historical BICR-adjudicated trial datasets before use; regulators accept the tool under a qualification pathway.

## Fields

- Kind: Idea
- Last checked: 2026-09-08
- Hypothesis: AI-assisted central reads will match BICR progression dates within one assessment interval in more than 95 percent of cases, at less than half the cost and with lower inter-reader variance, without changing trial conclusions on re-analysis.
- Rationale: Central imaging review is one of the largest fixed costs in phase 3 oncology trials and reader disagreement drives discordance between local and central PFS. Segmentation models now perform at expert level on common lesion types.
- Proposed test: Re-read the imaging archives of three completed phase 3 trials with the AI-assisted workflow and compare PFS hazard ratios and progression dates with the original BICR.
- Maturity: early-clinical
- Actor: engineering

## Sources

- Bottleneck evidence (Trial design, endpoints and cost): Davis et al., Availability of evidence of benefits on survival and quality of life of cancer drugs approved by EMA 2009-13 (BMJ 2017): https://doi.org/10.1136/bmj.j4530

## Connected records

- technologies: [AI in radiology](https://onco.cc/technologies/radiology-ai-screening/), [CT (computed tomography)](https://onco.cc/technologies/ct/)
- companies: [Aidoc](https://onco.cc/companies/aidoc/), [Lunit](https://onco.cc/companies/lunit/)
- terms: [Progression-free survival (PFS)](https://onco.cc/terms/pfs/), [RECIST](https://onco.cc/terms/recist/)
- bottlenecks: [AI that is built but not validated or deployed](https://onco.cc/bottlenecks/b-ai-validation/), [Trial design, endpoints and cost](https://onco.cc/bottlenecks/b-trial-design/)
- key papers: [Availability of evidence of benefits on overall survival and quality of life of cancer drugs approved by European Medicines Agency: retrospective cohort study of drug approvals 2009-13](https://onco.cc/key-papers/paper-davis-bmj/)
- roadmaps: [AI in oncology roadmap: pattern readers → foundation models → agents in the workflow](https://onco.cc/roadmaps/ai-oncology-roadmap/), [Trial modernisation roadmap: the randomised trial → platforms and adaptive designs → decentralised, pragmatic and always-on](https://onco.cc/roadmaps/trial-modernisation-roadmap/)

---
JSON: https://onco.cc/api/v1/entities/idea-tr1-ai-central-imaging-reads.json