Original article was published by on Communications of the ACM – Artificial Intelligence
By Ryerson University News
September 2, 2020
A three-year, multi-institutional study by 19 researchers at 12 institutions in five countries has yielded artificial intelligence (AI) models that can automate evaluations of stored red blood cell (RBC) quality that match or surpass expert assessment.
The investigators used 40,900 cell images to teach neural networks to classify RBCs into six categories; a fully-supervised machine learning algorithm agreed with human experts’ assessments 77% of the time (even human experts only agree 83% of the time).
Ryerson University’s Michael Kolios calls this achievement “a testament to how technology and science are now interconnecting to solve today’s biomedical problems.”
From Ryerson University News
View Full Article
Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA
No entries found