Artificial Intelligence Excels in Detecting Breast Cancer
A recent study has shown that artificial intelligence (AI) outperforms traditional methods in detecting breast cancer. Radiologists using AI successfully identified cancer cases at a rate of 6.7 per 1,000 screenings, representing a 17.6% increase compared to the 5.7 per 1,000 rate in the group relying solely on human evaluation.
The study, conducted by a team of researchers led by Dr. Alexander Katalinic from the University of Lübeck in Germany and published in Nature Medicine, highlighted that the use of AI contributed to an increased number of detected cases.
A total of 260,739 patients underwent dual evaluation, with images assessed by both radiologists and an AI system, while 201,079 patients were screened by radiologists alone. This division allowed for a direct comparison of radiologists’ performance with and without AI assistance.
The study revealed that when biopsies were performed following a suspected cancer diagnosis, the likelihood of finding cancerous cells in the biopsies increased to 64.5% in the group using AI, compared to 59.2% in the group without AI.
These findings indicate that the use of AI also helped improve diagnostic accuracy and more precisely identify cases requiring a biopsy.
