I just doing my master thesis and my theme is AI of medical image. I have a total of 1000 training images, but I should to clarify how many images I need to reduce the number of training images to. So I investigate how the number of training images affects the segmentation result by reducing the number of images by 100. However, if I reduce the number of images by 100, the training images will become more and more biased. How should I select the training images?
I searched a website and my friends. but i dont get good thing
If you have any idea or know interesting paper. please tell me!
Infrared is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.