
A health system in Canada that is reviewing thousands of mammograms originally interpreted with substandard medical displays was warned about the problem years previously, according to a news report.
A radiologist who formerly worked at Central Health in the province of Newfoundland and Labrador said she warned health authorities about problems in its mammography program in 2019, but she claimed her warnings were ignored, according to an article published by the Canadian Broadcasting Corporation (CBC).
Central Health announced in August that it was beginning a review of thousands of screening mammograms acquired at its sites for the past three years. The system said that these mammograms were interpreted with 3-megapixel medical displays, when established guidelines in Canada call for higher-resolution 5-megapixel displays to be used for mammography.
In the new CBC article, Dr. Jane Rendell said she told both the provincial government and regional health authority about issues in the health system's mammography program in 2019. Specifically, she said the breast screening program at the Central Newfoundland Regional Health Centre in the city of Grand Falls-Windsor wasn't accredited by the Canadian Association of Radiologists. That meant it hadn't gone through quality checks that might have found the problem.
The article said that in addition to Central Health, other health authorities in the region have discovered that mammograms were interpreted on 3-megapixel rather than 5-megapixel monitors and need to be reinterpreted.
The story also raised the issue of whether there are differences between 3-megapixel and 5-megapixel monitors that would be discernible to the human eye. The CBC quoted several radiologists who said it might not make a difference, but Rendell maintains that the higher resolution can help pick up small features like microcalcifications.
![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&fit=crop&h=100&q=70&w=100)







![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&fit=crop&h=112&q=70&w=112)










