Artificial intelligence-enabled detection of vascular perfusion defects on ventilation/perfusion (V/Q) scintigraphy for pulmonary embolism.
Artificial intelligence matched expert radiologists at detecting lung perfusion defects on nuclear scans while maintaining low false-alarm rates.
This retrospective diagnostic model validation study examined artificial-intelligence detection of perfusion defects on ventilation-perfusion scintigraphy. The best-performing model matched human readers and retained low false-positive rates on a separate low-likelihood holdout cohort.
What the study was
- Study design
- Retrospective diagnostic model validation study
- Population
- 2118 patients with ventilation-perfusion scans plus a 430-patient holdout set
- Sample size
- 2548
- Category
- Diagnostics
- Maturity
- Validated
- Journal
- Physics in medicine and biology
Why it surfaced
Retained as a standard-priority retrospective diagnostic model validation study after PubMed abstract review; the score reflects clinical relevance, design strength, novelty, and population need under the v1.3 conservative rubric.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.