VIDEO: Potential of AI in breast cancer diagnostics

Editor’s note: This is an automatically generated transcript. Please notify editor@healio.com if there are concerns regarding accuracy of the transcription.

So it's really exciting to see artificial intelligence move into breast cancer therapy and diagnostics. I think it's still in its infancy, but the potential that I think we are envisioning is really exciting. So for example, right now AI is often integrated into radiologic reads. So, with mammography reads, a lot of mammographers use an AI tool that will help point out areas of abnormality on the mammogram, but then there's a human being who's doing a read of it after getting that information. And so it's really, I think, able to allow for more sophisticated reads that hopefully will pick up findings more accurately than previously. Another example would be with pathology. It's becoming complicated in breast cancer to understand target expression. So for example, one area that we're struggling with is trying to understand how much of the HER2 receptor is present on the cancer cell surface. And, when a pathologist reads it, they can make estimates of what percent of cells have what level of HER2 expression, but to get a real accurate percentage would take a machine algorithm to do that. And so there are lots of companies now that have started using machine algorithms with AI technology to really be able to give you a quantitative assessment of that HER2 read. And so I think, again, that's certainly not standard and not what we're doing in all cases, but I think it does show that there's a movement towards being able to do that. And hopefully in the future, these kinds of technologies will be become more standard so that we'll be able to give more accurate diagnoses and assessments for our patients.