Medical Image Quality Assurance using Deep Learning
Dženan Zukić, Anne Haley, Curtis Lisle, James Klo, Kilian M. Pohl, Hans J Johnson, Aashish Chaudhary
We present an open-source web tool for quality control of distributed imaging studies. To minimize the amount of human time and attention spent reviewing the images, we created a neural network to provide an automatic assessment. This steers reviewers' attention to potentially problematic cases, reducing the likelihood of missing image quality issues. We test our approach using 5-fold cross validation on a set of 5217 magnetic resonance images.
Wednesday 6th July
Poster Session 1.2 - onsite 11:00 - 12:00, virtual 15:20 - 16:20 (UTC+2)