Lunit INSIGHT MMG at Capio S:t Göran Hospital
Lunit INSIGHT MMG used for AI as an independent reader in double-reading breast cancer screening, fully replacing one of the two human readers — vendor-reported deployment at Capio S:t Göran Hospital.
- What this means:
- The source has been read and our reading of it confirmed — the classifications that let this record be compared with others, the AI product, and the clinical use. That is a check of our work, not a judgement on the AI.
- What happens next:
- Nothing, unless this use case is taken up for assessment. Then the AI product and its clinical use are examined together in seven sections — purpose, evidence, CE conformity, legal duties, ethics, safety, and what a hospital must have in place — each signed by a named institution.
- Submitted
- 2Curated
- 3Maturing
- 4Validated
- 5Scaled
- Clinical problem
- Detect a condition
- Point of care
- Diagnosis
- Nature of AI output
- A classification
- Clinical specialty
- Radiology
- Care setting
- Hospital — outpatient
- Patient population
- Not specified in source (vendor case study)
- Intended use
- Lunit INSIGHT MMG used for AI as an independent reader in double-reading breast cancer screening, fully replacing one of the two human readers — vendor-reported deployment at Capio S:t Göran Hospital.
Full study description
AI as an independent reader in double-reading breast cancer screening, fully replacing one of the two human readers. Vendor-reported deployment at Capio S:t Göran Hospital. Vendor case study reports: "Cancer detection rates increased by 15%, rising from 4.8 (traditional double reading) to 5.5 per 1,000 screened". Vendor case study reports: "11% Decrease in Recall Rates: Recall rates dropped from 2.8% to 2.5%". Vendor case study reports: "Reading time decreased by over 36%". Not independently verified. Source: https://www.lunit.io/en/media-hub/real-world-validation-lunit-ai-proven-successful-in-1-year-breast-cancer-screening-deployment/
- AI technique
- Deep learning, Computer vision
- Input data
- Medical imaging
- Output type
- Classification
- Autonomy level
- Informs a human (advisory)
- Model provenance
- Vendor proprietary
- Status
- Active deployment
- Country
- Sweden
- Deployment date
- 1 January 2024
- Data identifiability
- Not stated in the source
Similar deployments
AIEvidence records
Studies and evaluations attached to this use case.
implementation_study
cancer_detection_rate_per_1000: 5.5
Population: Post-implementation with AI as one of two readers, vs 4.8 per 1,000 with traditional double reading (a 15% increase). Capio S:t Goran, July 2023-March 2024. Vendor-reported (Lunit); corroborated by the hospital's own site confirming routine clinical use.
Sourceimplementation_study
recall_rate_percent: 2.5
Population: Down from 2.8% pre-implementation. Same cohort/period. Vendor-reported.
Sourceimplementation_study
false_positive_rate_percent: 78
Population: Down from 89.6% pre-implementation. Same cohort/period. Vendor-reported.
Sourceimplementation_study
reading_time_reduction_percent: 36
Population: Reading time decreased by over 36%. Same cohort/period. Vendor-reported.
Source
Contributors
- Deploying organisation
- Capio S:t Göran Hospital · Hospital / health system · Sweden
- AI vendor
- Lunit Inc. · South Korea
- Product name
- Lunit INSIGHT MMG
- Source record
- Vendor case study