00:00 This might look like a normal microscope, but it's being driven by artificial intelligence.
00:07 So for ten years we've been working on a system that can image these huge pathology slides
00:13 so we can then apply AI to the diagnosis.
00:17 Humans no longer have to peer into the machine and search for diseases.
00:22 Once we used to look at 100 cells to classify those cells, now we can look at 1,000 cells
00:28 and you get a much better outcome.
00:30 The AI is used during the image capture. The system actually knows what it's looking for,
00:35 it knows what diagnostic cells look like.
00:37 Robots already help out with menial pathology tasks like sorting and transporting samples,
00:44 but now AI can help diagnose diseases, identify infections.
00:50 An experienced, trained scientist can tell the AI, look, that's that pattern and the
00:56 AI will remember that. So the next time it sees a digital image with that pattern it'll
01:01 say, oh, that's X, and so it'll classify.
01:05 Currently we support 17 different tests that are run on a 24/7 basis.
01:10 The AI project is a collaboration between the University of Queensland and Sullivan-Nicoletti's
01:15 Pathology.
01:16 Still within human control. This is not a robot that's doing this. This is a computer
01:22 program that's been trained by an expert and will be continually reviewed by that expert
01:29 as well.
01:30 A human signs off on everything, but we make it so that it's much easier for the human
01:34 to sign off quickly.
01:35 The hope is that technology can be adopted by other labs.
01:38 We've got an AI algorithm that accurately diagnoses them. I believe it is a world first.
01:43 [BLANK_AUDIO]
Comments