Evaluating Knowledge Transfer In Neural Network for Medical Images

Deep learning and knowledge transfer techniques have permeated the field of medical imaging and are considered as key approaches for revolutionizing diagnostic imaging practices . However, there are still challenges for the successful integration of deep learning into medical imaging tasks due to a lack of large annotated imaging data… To address this issue, we propose a teacher-student learning framework to transfer knowledge from a carefully pre-trained convolutional neural network (CNN) teacher to a student CNN . We also examine the proposed network’s behavior on the convergence and regularization of the student network during training . The performances of the CNN models are evaluated on three medical imaging datasets including Diabetic Retinopathy, CheXpert, and ChestX-ray8 . Our results indicate that the teacher-Student learning framework outperforms transfer learning for small imaging datasets. We also demonstrate a clear advantage to favoring teacher-worker learning framework for cross-domain knowledge transfer in the medical imaging setting compared to other knowledge transfer methods such as transfer learning. In addition to small training data size, we also demonstrate to favoring a teacher and student network, we say, it also demonstrates to favoring the teaching-learning framework for large training data sizes and to reduce overfitting when the dataset is small. We observe that the teaching network holds a great promise not only to improve the performance of diagnosis but also to reduce the overfitting and to improve overfitting of the data size. We conclude that

Links: PDF - Abstract

Code :


Keywords : learning - imaging - transfer - network - teacher -

Leave a Reply

Your email address will not be published. Required fields are marked *