--- title: >- Capital Region of Denmark mammography screening programme: AI-triaged single versus double reading of screening mammograms (Transpara) slug: denmark-mammography-triage stable_id: fbf20734fb447a59 company: Capital Region of Denmark mammography screening programme function_code: healthcare_screening pattern_codes: - exception_based_operations - decision_right_transfer - creator_to_judge evidence_strength: verified publication_tier: showcase freshness: current reviewed_at: '2026-08-23' updated_at: '2026-08-22' source_quality_summary: 2 independent, 1 peer reviewed; publication outcomes are verified. caveat_summary: >- Observational sequential cohorts, not randomised. Interval cancers were not yet available. Screening interval lengthened after AI (median 845 to 993 days), which the authors stratified and argue does not fully explain the detection gain. Screen-detected cancers only, with ≥180 days follow-up. DCIS share rose (15.1% to 20.4%), so overdiagnosis remains an open question pending interval-cancer follow-up. False-positive rate also fell (2.39% to 1.63%) and PPV rose (22.6% to 33.6%). Not a randomised comparison. collections: [] bundle_version: 1.0.0 bundle_fingerprint: sha256:c23c6cc2b88153d008ea8fda928f632ce0011fc2d4c5036672a16e5d895bab93 canonical_url: https://brianletort.ai/transformations/denmark-mammography-triage --- # Capital Region of Denmark mammography screening programme: AI-triaged single versus double reading of screening mammograms (Transpara) That every screening mammogram must be independently double-read by two radiologists to keep cancer detection and recall rates acceptable. Function: Healthcare screening. Patterns: Exception-based operations; Decision-right transfer; Creator to judge. Evidence: verified. Freshness: current. Reviewed: 2026-08-23. Updated: 2026-08-22. Source quality: 2 independent, 1 peer reviewed; publication outcomes are verified. ## Before 1. **Radiographer** — Acquires four-view full-field digital mammograms under Danish screening guidelines. (control: Acquisition protocol; PACS routing) 2. **Two independent breast radiologists** — Each reads every examination; the second reader is always a senior radiologist. (control: Independent double reading; European/Danish screening guidelines) 3. **Consensus meeting or third-reader arbitration** — Resolves recall disagreements. (control: Consensus or arbitration) ## After 1. **Transpara AI (ScreenPoint Medical v1.7.1)** — Assigns an examination malignancy score (1–10) and, for higher-risk exams, lesion marks. (control: Score threshold (≤5 until 2022-05-03, then ≤7) set by the programme with continuous quality monitoring) 2. **Senior breast radiologist (single-read lane)** — Reads likely-normal examinations knowing they are single-read; may still send a recall to consensus. (control: Human recall authority retained; 0.8% of single-read exams were recalled against the AI-normal tag) 3. **Two radiologists with AI decision support (higher-score lane)** — Independently double-read remaining examinations and consult AI marks after their own evaluation. (control: Readers blinded to each other; instructed not to use AI marks as the first look) ## Decision rights The programme sets the score threshold that drops the second human reader. Radiologists retain recall/no-recall authority, including the right to override an AI-normal tag. ## Exception path Unprocessed exams (1.5%, mostly early infrastructure failures or implants) revert to double reading. Single-read recalls go to consensus or arbitration. Continuous quality control justified raising the threshold from 5 to 7. ## Outcomes - **Radiologist screening-read workload** (verified): Two reads per examination (double reading of the entire cohort) → 33.5% fewer reads (38,977 of 116,492 possible reads saved) because 66.9% of exams were single-read. Observational sequential cohorts, not randomised. Interval cancers were not yet available. Screening interval lengthened after AI (median 845 to 993 days), which the authors stratified and argue does not fully explain the detection gain. - **Cancer detection rate** (verified): 0.70% (423 of 60,751) → 0.82% (480 of 58,246); P = .01. Screen-detected cancers only, with ≥180 days follow-up. DCIS share rose (15.1% to 20.4%), so overdiagnosis remains an open question pending interval-cancer follow-up. - **Recall rate** (verified): 3.09% (1,875 of 60,751) → 2.46% (1,430 of 58,246); 20.5% relative reduction; P < .001. False-positive rate also fell (2.39% to 1.63%) and PPV rose (22.6% to 33.6%). Not a randomised comparison. ## Executive lesson The Capital Region did not give radiologists a faster viewer. It changed who must look. AI became the first sort, senior radiologists became the only readers on the likely-normal majority, and recall stayed human. The measured result is fewer reads, fewer recalls, and more screen-detected cancers in a live national-style programme. That is the MASAI trial's missing piece: standing practice, not a protocol. ## Anti-pattern Citing MASAI or ScreenTrustCAD as if they were BAU operations. Those are trials. This case is the programme that actually changed the reading protocol. ## Questions for leaders - Where do we still insist on two humans for work a model can already sort by risk? - Who owns the threshold that drops a reviewer, and how often is it re-validated? - What override rate would tell us the single-read lane is unsafe? ## Sources - [Early Indicators of the Impact of Using AI in Mammography Screening for Breast Cancer](https://doi.org/10.1148/radiol.232479) — Radiology (RSNA) - [PubMed record for Lauritzen et al., Radiology 2024](https://pubmed.ncbi.nlm.nih.gov/38832880/) — PubMed / NLM - [AI successfully helps detecting more breast cancers around Danish capital](https://science.ku.dk/english/press/news/2024/ai-successfully-helps-detecting-more-breast-cancers-around-danish-capital/) — University of Copenhagen Faculty of Science