LO
DeploymentCognitive automation2020-09-30
At a cancer hospital, AI built to flag incidental pulmonary embolism on routine CT cut radiologists' missed rate for that finding from 44.8% to 2.6%, across 11,736 scans
Radiologistoccupation page →
Event date / reported
2020-09-30 · reported 2023-04-20
Evidence stage
DeploymentAn employer has put it into production. Can move the baseline — weighted by scale and how similar the setting is.
Tasks this bears on
Catching what nobody asked about
The scan was ordered for a back injury; the thing that matters is in the corner of the image and has nothing to do with the back.
Still human-led✓ Evidence-backed
Where this applies
Radiologists at a cancer hospital evaluating commercial AI used to prioritise incidental pulmonary embolism on routine oncology CT scans, across three periods from April 2019 to September 2020: routine workflow, human triage, and AI-prioritised worklists. Of 11,736 scans in 6,447 patients, the software reached 91.6% sensitivity and 99.7% specificity; radiologists' missed rate of incidental embolism fell from 44.8% (47 of 105 scans) without AI to 2.6% (one of 38) with it, and time to notification fell sharply. The software looks for this one finding only; the vendor supplied it, and one author has ties to other AI vendors.
What this means
A known incidental finding can be turned into a question a machine is asked — and when it was, radiologists stopped missing most of them. But the tool looks for one thing; the finding nobody has built a model for is still caught, or missed, by the person reading the whole scan.
What it does not yet show
One hospital, one pre-specified finding, with the vendor's software; it does not show AI catching findings nobody anticipated.
What you can check
Open Topff et al., Radiology: Cardiothoracic Imaging 2023 (PMC10141443), and find "missed rate of IPE was significantly reduced from 44.8%".
Does it change the assessment?
No. The impact index is never moved by a single event. What this record did: the 1 linked task judgement above now rest on evidence instead of inference.
Source
Topff L. et al. (Netherlands Cancer Institute) — "Artificial Intelligence-assisted Workflow Prioritization of Incidental Pulmonary Embolism on Routine CT", Radiology: Cardiothoracic Imaging 2023;5(2):e220163 (published online 20 April 2023) · verified 2026-09-28 · Claude (VOLO agent) · interpreted 2026-09-28 · Claude (VOLO agent)
Primary source — published by the party that did this, or the authority of record. No co-signature needed.