Updated:
Keep
The artificial intelligence (AI) it is a promising tool for the detection of breast cancer in screening programs as it could speed up diagnoses, according to a study published in Radiology.
Mammograms done through breast cancer screening programs create a significant workload for radiologists. And, while the technology has shown promising results for cancer detection, evidence regarding its use in real-world screening settings is limited.
In this study, the largest of its kind to date, Norwegian researchers led by Solveig Hofvindthe Norwegian Cancer Registry in Oslocompared the performance of a commercially available AI system with routine independent double reading as performed in a country population-based screening program.
The research was based on almost 123,000 exams conducted on more than 47,000 women at four treatment facilities BreastScreen Noruegathe nation’s population-based screening program.
The data set included 752 cancers detected at screening and 205 interval cancers, or cancers detected between rounds of screening. The artificial intelligence system predicted cancer risk on a scale of 1 to 10, with 1 representing the lowest risk and 10 the highest risk. A total of 87.6% (653 of 752) of screen-detected cancers and 44.9% (92 of 205) of interval cancers had the highest AI score of 10.
The researchers created three thresholds to assess the performance of the AI system as a decision-making tool. Using a threshold that reflects the radiologists’ individual average rate of positive interpretation, the proportion of screen-detected cancers that were not selected by the AI system was less than 20%. While the AI system worked well, the study’s reliance on retrospective data means more research is needed.
“In our study, we assumed that all cancer cases selected by the AI system were detected,” says Hofvind. “This might not be true in a real detection environment. However, given that assumption, AI will likely be of great value in the interpretation of screening mammograms in the future.”
The results showed favorable histopathologic features associated with a better prognosis for screen-detected cancers with low versus high AI scores. Opposite results were seen for interval cancers. This may indicate that interval cancers with low AI scores are true interval cancers that are not visible on screening mammograms.
“Based on our results, we expect AI to be of great value in the interpretation of screening mammograms in the future,” the researchers note.
Although more studies are needed before the clinical implementation of AI in breast cancer detection, the study results help establish a foundation for future research, including prospective studies.
See them
comments