Mia
Commercial AI reader for screening mammography, evaluated as an additional reader in the UK's double-reading pathway; the GEMINI prospective study in Aberdeen reported extra cancers found when Mia flagged cases the readers had cleared.
At a glance
What it does
Mia is an AI system that supports the reading of mammograms in breast cancer screening programmes, acting as an additional reader alongside radiologists. It highlights examinations and findings that may warrant further review.
Notable uses in oncology
In a prospective NHS Grampian evaluation in Scotland covering 10,889 women, the AI-supported workflow helped detect 12% more cancers than routine practice — mostly invasive, high-grade tumours — with no increase in unnecessary recalls; modelling suggested a workload reduction of up to 30%.
Tasks, data types and cancers
Architecture
Training data
Not disclosed by the manufacturer. Checked on 2026-09-06: kheironmed.com now redirects (HTTP 301) to deephealth.com/kheiron-medical-technologies/, which states that Kheiron Medical Technologies became part of DeepHealth (a wholly owned subsidiary of RadNet) in 2024. That page names no training dataset, no number of mammograms or patients, no source institutions and no architecture; it covers company history, CE marking (2019) and clinical deployments instead.
Evaluation
| Benchmark / dataset | Metric | Value | External validation | Source |
|---|---|---|---|---|
| GEMINI — prospective evaluation in NHS Grampian screening service, 10,889 women (mammograms read 27.02–05.10.2023) | cancer detection rate / recall rate / reader workload | Primary AI workflow: cancer detection +10.4% (1 additional cancer per 1,000 screened), recall rate maintained (0.8% reduction), reader workload reduced by up to 31%. Where AI recommended recall but standard double reading did not, additional human review identified 11 extra cancers. Alternative simulated workflows reached workload savings of up to 36%. All participants received routine care; the AI did not act alone. | yes | Source |
How to run
Not yet documented on this card.
Regulatory status and intended use
The manufacturer's page states that Mia (Mammography Intelligent Assessment) received CE marking in Europe in 2019; no FDA clearance or UKCA marking is stated there. Kheiron Medical Technologies became part of DeepHealth (a wholly owned subsidiary of RadNet) in 2024. Verified 2026-09-06: the owner's page refers to Mia in the past tense ("Kheiron was known for its pioneering product"), so current commercial availability and the scope of the certificate should be confirmed with the vendor before any clinical use.
Regulatory status is quoted from the source linked above and can change. Research-use-only models must not be used for clinical decisions.
Limitations and bias
- Closed model; evidence is from deployment studies rather than open benchmarks.
- Results depend on the screening programme's double-reading design.
Sources
- GEMINI study — Prospective evaluation of artificial intelligence integration into breast cancer screening in multiple workflow settings, Nature Cancer 2026;7(3):484-493
- University of Aberdeen — GEMINI study announcement (2024; superseded figures)
- DeepHealth — Kheiron Medical Technologies (manufacturer page)
This page is educational — it is not medical advice and does not replace consultation with an oncologist. Diagnostic and treatment decisions are made solely by specialist physicians.