AI-assisted breast cancer screening: Promising results but many questions remain

The research group “Artificial Intelligence in Breast Cancer Screening Using Mammography” was prepared by HTA syd at Skåne University Hospital and experts in the field. Together they looked at the best available research on the use of AI in breast cancer screening using mammography, to see if it was possible to draw conclusions about effectiveness and outcomes.

The clustering provides some answers about performance in breast cancer detection, but it also shows that many questions about the impact of AI in breast cancer screening remain unanswered. For example, it is not possible to draw any conclusions about whether cancers detected using AI are more or less dangerous, compared to the current routine, says Maria Madestam, an expert in research synthesis and chief medical officer of mammography at Unilab in Helsingborg. . .

Artificial intelligence can save time

The compilation shows that using AI can free up radiologists to work on other things. But using AI also involves new costs, and calculations suggest it is not cheaper for regions, at least not in the short term. It is therefore important to ensure that AI leads to health gains or other healthcare benefits that can justify these cost increases.

facts

• HTA stands for Health Technology Assessment and is a method for assessing the state of knowledge of questions raised by healthcare processes.
• The research has been compiled in collaboration between experts in various specialties from the Healthcare District South and HTA South at Skåne University Hospital.
• The research was compiled by Unilabs, which is responsible for screening for breast cancer using mammography in the Scania region.