OnCo
ideasIdea

No clinical claims for imaging-derived biomarkers without phantom and standards compliance

Thousands of papers extract 'radiomic' features from scans to predict outcomes, but the features change with scanner settings. Journals should require standard compliance before any clinical claim is made.

Radiomic features vary with acquisition, reconstruction and software; the Image Biomarker Standardisation Initiative (IBSI) defined reference values and benchmark datasets, but most published radiomics studies do not report compliance or test-retest reproducibility. Journals and funders requiring IBSI compliance, phantom-based feature stability testing and external validation for any radiomic biomarker claiming clinical relevance would cut the noise.

Hypothesis
Adopting the requirement will reduce the number of published radiomic signatures by more than half and increase the fraction that replicate in external cohorts from a small minority to a majority.
Rationale
The field's replication crisis is well documented in systematic reviews; the standards exist and are unenforced.
What would test it
Two radiology journals adopt the requirement; compare external validation rates of signatures published before and after.
Maturity
early clinical
Who has to act
research
Cost to try
Small (under $1M)
Years to first evidence
2
Bottlenecks it attacks

Connected

7top

Pages like this

not linked directly; found by shared links

technologies

3
CT (computed tomography)MRIPET/CT

companies

2
AidocLunit

bottlenecks

2
AI that is built but not validated or deployedBiomarkers are not validated or standardised