Our paper titled “Myelin Detection in Fluorescence Microscopy Images Using Machine Learning” is published online at the Journal of Neuroscience Methods. To the best of our knowledge, this is the most thorough evaluation of machine-learning based methods for myelin identification up to date. We tested 23 methods including our customized-CNN. Both the customized-CNN and Boosted Trees methods were robust and highly effective in identifying myelin sheath on fluorescence microscopy images. You can read the paper here.
I would like to thank our collaborators Behcet Ugur Toreyin, Abdulkerim Capar and Umut Engin Ayten for their excellent work. Sibel Cimen Yetis did a fantastic job with help of Dursun Ali Ekinci.
Stay tuned for our results on myelin quantification i.e. detection and measurement of axons and myelin under different culture conditions using machine-learning.
