The artificial intelligence system was more accurate than human doctors in distinguishing between cancers.
Analysing breast biopsies can be incredibly challenging, even for highly experienced specialist practitioners.
But help may soon be on the way: a team based at the University of California, Los Angeles has developed an artificial intelligence system that could help to read biopsies more accurately.
“Medical images of breast biopsies contain a great deal of complex data and interpreting them can be very subjective,” said Dr. Joann Elmore. “Sometimes, doctors do not even agree with their previous diagnosis when they are shown the same case a year later.”
The team fed 240 breast biopsy images into a computer, training it to recognise patterns associated with several types of breast lesions, from harmless to cancerous, under the supervision of three expert pathologists.
They then tested the system by comparing its reading to a set of diagnoses made by 87 practising pathologists. While the system performed slightly below the human doctors in differentiating cancer from non-cancer cases, it outperformed them when challenged with distinguishing between two difficult-to-distinguish cancers: ductal carcinoma in situ (DCIS), a non-invasive type of breast cancer, and breast atypia, a cell abnormality associated with a higher risk for cancer.
“These results are very encouraging,” Elmore said. “There is low accuracy among practicing pathologists in the U.S. when it comes to the diagnosis of atypia and ductal carcinoma in situ, and the computer-based automated approach shows great promise.”
The researchers are now working on training the system to diagnose melanoma.