research
Three Readers Capture Most of the Gaze Signal
Every multi-reader panel improved chest X-ray severity prediction over the no-gaze baseline, while no individual reader reached significance on the main endpoint.
Summary
Every multi-reader panel improved chest X-ray severity prediction over the no-gaze baseline, while no individual reader reached significance on the main endpoint.
GazeRIB asks whether radiologists' visual attention can supervise a model more reliably when it is pooled across readers. Eight experts repeatedly read the same 80 chest X-rays; the study compared individual readers with three-, five- and eight-reader panels across seven architectures and ten seeds, then evaluated 910 models on a separate 2,897-image cohort. Every panel improved macro F1 by 5.33 to 7.47 percentage points over the no-gaze baseline, while no individual reader reached significance on that endpoint. Three readers captured 73 to 80 percent of the eight-reader gain across macro F1, ROC-AUC and precision-recall area. External transfer was mixed: panels improved ROC-AUC on SIIM-FISABIO-RSNA, but not on MIMIC-Eye. The result supports panel composition as an experimental variable, not a claim that gaze guidance will transfer to every clinical setting.
Why it matters
Every multi-reader panel improved chest X-ray severity prediction over the no-gaze baseline, while no individual reader reached significance on the main endpoint.
Limits and context
- External transfer was mixed: panels improved ROC-AUC on SIIM-FISABIO-RSNA, but not on MIMIC-Eye.
- The result supports panel composition as an experimental variable, not a claim that gaze guidance will transfer to every clinical setting.
Key claims
Every multi-reader panel improved chest X-ray severity prediction over the no-gaze baseline, while no individual reader reached significance on the main endpoint.
Qualification: External transfer was mixed: panels improved ROC-AUC on SIIM-FISABIO-RSNA, but not on MIMIC-Eye.
Evidence: source-2026-11-02-001
Sources
- arXiv preprint 2610.42152, original v1arXiv · primary research
Corrections
No corrections have been recorded for this story.