Recently, a large-scale clinical study involving 100,000 women in Sweden showed that introducing artificial intelligence (AI) technology in breast cancer screening can effectively reduce the rate of missed cancers by 12% in the following years. This groundbreaking finding, published in The Lancet, provides the strongest evidence so far for the use of AI to assist radiologists and improve early diagnosis.

In this trial conducted from 2021 to 2022, participants were randomly divided into an AI-assisted group and a standard two-person manual reading group. The results showed that the AI group detected more than 80% of cancer cases during the screening phase, significantly higher than the 74% detected by the control group. More encouragingly, AI technology demonstrated stronger capabilities in identifying highly threatening invasive cancers, with a significant reduction in the rate of missed diagnoses for these types of cancers.

The AI system used in the study assessed risk based on mammograms, assigning low-risk cases to single-reader review and high-risk cases to double-reader review with marked suspicious areas. The research team from Lund University in Sweden pointed out that this model not only improves diagnostic accuracy but also significantly reduces the workload of radiologists, alleviating the pressure on medical resources.

Currently, breast cancer is the leading cause of death among women aged 35 to 50, with more than 2 million new cases diagnosed worldwide each year. Although AI has shown great potential, the researchers emphasized that it is not intended to replace professional doctors. At this stage, breast cancer screening still requires at least one doctor to make the final judgment with AI support.

Although the study results are promising, experts have also offered rational reminders. The head of the UK Cancer Research Centre stated that since the study was conducted at a single center, further multi-regional and multi-institutional studies are needed to verify whether AI can truly improve patient survival rates in the long term.

Dr. Kristina Lång from Lund University advised that medical institutions must be cautious when introducing AI tools. Continuous monitoring of AI tools is necessary to ensure they provide stable and reliable data support across different regions and screening programs. With the expected increase in cancer diagnoses in the coming years, AI and other technological innovations will become key tools in optimizing the healthcare system and saving more lives.