AI can improve breast cancer diagnosis, according to new study in Germany involving more than 461,000 people … [+]
Artificial intelligence (AI) can improve breast cancer detection and reduce the workload for doctors, according to a new study of more than 461,000 women. The research published in the journal Naturopathy looked at the integration of an AI tool as part of a national screening program for women without symptoms of breast cancer in Germany.
The majority of previous studies using AI to diagnose cancer have been conducted using existing data after people have actually been diagnosed. This is a necessary step to validate the use of new diagnostic methods, essentially to see how the AI algorithm would have fared in terms of accuracy and sensitivity if it had been used, without any risk to the patients. But the new study used AI prospectively in a national screening study in Germany of women between the ages of 50 and 69 who had no symptoms, in the hope of detecting early-stage cases of breast cancer.
The women were screened between 2021 and 2023 and divided into two groups. In one group, screening data were analyzed by two expert radiologists. In the other group, two radiologists also analyzed the data, but at least one of them also used an AI tool to help.
A total of 2,881 women were diagnosed with breast cancer, with the AI group having a 6.7% higher detection rate. When adjusting for possible variables in the groups, such as the age of women diagnosed, this difference increased to 17.6% higher diagnoses in the AI-assisted group.
“We actually wanted to show with the study that AI reporting is equivalent to human reporting,” explains Alexander Katalinic, MD, research leader and director of the Institute for Social Medicine and Epidemiology at the University of Lübeck and the UKSH, Lübeck Campus in Germany out. “But the results positively surprised us: AI actually significantly improves the detection rate of breast cancer,” said Katalinic press release.
The study also looked at “false positives” – when a suspicious scan is labeled as potentially cancerous, prompting further investigation, when in fact it is not. Cancer detection with AI has sometimes historically had a higher false positive rate, and these can lead to significant suffering for patients and additional costs to healthcare systems. Both AI-assisted and unassisted radiologists had a similar rate of false-positive results.
“The PRAIM study shows the enormous potential of AI to improve screening programs worldwide,” says Stefan Bunk, CTO of the AI company Vara that designed the AI tool. “This evidence will take the discussion about integrating AI into healthcare systems to a new level,” Bunk added.
There is hope that AI tools can eventually reduce the workload of doctors involved in diagnosing cancer, improve accuracy and reduce the number of false positives. Most studies to date have been conducted with physicians still validating all AI results to limit the risk of additional harm to patients due to missed diagnoses or false positives.
“We hope that the higher hit rate with AI can further improve the prognosis for women with breast cancer. We will investigate this in the future,” Katalinic said.