LibraryFamily 1 · Diagnostic Imaging & Pathology
AI second reader for breast cancer screening
ValidatedActive deploymentEmerging evidence
Formally validated with a published Assurance Pack.
Acts as an independent second reader of screening mammograms to support the double-reading workflow and reduce reader workload without lowering cancer detection.
Plain-language summary
AIChoose an audience and generate a tailored summary on demand. The AI uses only what is on this page.
Clinical context
- Clinical problem
- Detect a condition
- Point of care
- Diagnosis
- Nature of AI output
- A classification
- Clinical specialty
- Radiology
- Care setting
- Hospital — outpatient
- Patient population
- Women aged 50–70 attending routine NHS Breast Screening Programme mammography.
- Intended use
- Acts as an independent second reader of screening mammograms to support the double-reading workflow and reduce reader workload without lowering cancer detection.
Technology
- AI technique
- Deep learning, Computer vision
- Input data
- Medical imaging
- Output type
- Classification
- Autonomy level
- Informs a human (advisory)
- Model provenance
- Vendor proprietary
- Model version
- Mia v2.1
- Built on a general-purpose model
- No
Deployment
- Status
- Active deployment
- Country
- United Kingdom
- Deployment date
- 1 March 2024
- Sites
- 3
Regulatory & governance
- EU AI Act risk tier
- High-risk
- High-risk basis
- Annex I — medical device
- Medical device
- Yes
- EU MDR class
- Class IIb
- CE marking
- CE marked
- FDA status
- Not submitted
- ISO 14971 risk class
- Medium
- GDPR processing basis
- Public interest
- GDPR DPIA
- Completed
- Data identifiability
- Pseudonymised
- Explainability method
- Post-hoc
- Human oversight model
- Every flagged case is reviewed by a qualified radiologist before recall; AI never recalls a patient autonomously.
NICE evidence standards
- ESF tier
- Tier C — treat / diagnose / calculate risk
- Evidence category
- Category 3
Performance summary
- Headline metric
- Sensitivity
- Value
- 0.91
- Subgroup performance assessed
- Yes
- Known bias signals
- Slightly lower sensitivity reported for dense breast tissue (BI-RADS density C/D); monitored as part of the validation review.
Similar deployments
AILooking for close matches in the Library…
Assurance Pack
Validation artefact produced by the AIH Lab review panel.
- Criteria version
- 2026.1
- Result
- PASSED
- Issued
- 1 February 2024
- Panel chair
- —
Evidence records
Studies and evaluations attached to this use case.
Retrospective validation
Specificity: 0.89
Population: NHS screening cohort, 120k mammograms
Retrospective validation
Sensitivity: 0.91
Population: NHS screening cohort, 120k mammograms
Contributors
- Deploying organisation
- [demo] Oxford University Hospitals NHS FT · Hospital / health system · United Kingdom
- AI vendor
- [demo] Kheiron Medical Technologies · United Kingdom
- Product name
- Mia