2026-03-15T21:12:40Z https://tokushima-u.repo.nii.ac.jp/oai

oai:tokushima-u.repo.nii.ac.jp:02011943 2025-04-28T06:24:55Z 1713853213384:1713853295607

Looking Closer to the Transferability Between Natural and Medical Images in Deep Learning Looking Closer to the Transferability Between Natural and Medical Images Rufaida, Syahidah Izza Putra, Tryan Aditya Leu, Jenq-Shiou 宋, 天 ソウ, テン Song, Tian 宋 ソウ Song 天 テン Tian 片山, 貴文 カタヤマ, タカフミ Katayama, Takafumi 片山 カタヤマ Katayama 貴文 タカフミ Takafumi open access This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/ Data augmentation medical image dataset meta-learning natural images dataset transfer-learning Transfer-learning has rapidly become one of the most sophisticated and effective techniques in dealing with medical datasets. The most common transfer-learning method uses of a state-of-the-art model and its corresponding parameters as the starting point for new tasks. Recent studies have found that transfer-learning between medical and natural images has minimal advantages, attributed to their different characteristics, even with sufficient data and iterations. This study employs a meta-learning technique, building upon the traditional transfer learning approach, to explore the potential of natural tasks as a starting point for analyzing medical images. In addition, this study investigates the performance of transferring the searched augmentation from natural to medical images. Several studies proposing search algorithms for data augmentation argue that the augmentation techniques can be effectively transferred across different datasets. The results revealed that the transferability between natural and medical images leads to reduced performance owing to the characteristic difference between medical and natural searched augmentation. IEEE 2023-07-28 eng journal article VoR 399068 https://tokushima-u.repo.nii.ac.jp/records/2011943 https://doi.org/10.1109/ACCESS.2023.3299819 10.1109/ACCESS.2023.3299819 21693536 IEEE Access 11 79838 79850 https://tokushima-u.repo.nii.ac.jp/record/2011943/files/access_11_79838.pdf application/pdf 1.31 MB 2024-06-14