Limited number of cases may yield generalizable models, a proof of concept in deep learning for colon histology

• Главные авторы: Lorne Holland (Автор), Dongguang Wei (Автор), Kristin A Olson (Автор), Anupam Mitra (Автор), John Paul Graff (Автор), Andrew D Jones (Автор), Blythe Durbin-Johnson (Автор), Ananya Datta Mitra (Автор), Hooman H Rashidi (Автор)
• Формат: ‌
• Опубликовано: Elsevier, 2020-01-01T00:00:00Z.
• Предметы: article
• Online-ссылка: Connect to this object online.

Background: Little is known about the effect of a minimum number of slides required in generating image datasets used to build generalizable machine-learning (ML) models. In addition, the assumption within deep learning is that the increased number of training images will always enhance accuracy and...