TY - GEN
T1 - DiRA
T2 - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
AU - Haghighi, Fatemeh
AU - Taher, Mohammad Reza Hosseinzadeh
AU - Gotway, Michael B.
AU - Liang, Jianming
N1 - Funding Information:
We propose DiRA, the first SSL framework that unites discriminative, restorative, and adversarial learning in a unified manner. The key contribution of our DiRA arises from the insights that we have gained into the synergy of these three SSL approaches for collaborative learning. Given DiRA’s generalizability, we envisage it will take a fundamental step towards developing universal representations for medical imaging. Our DiRA achieves remarkable performance gains, though we fixed the restorative learning tasks in all experiments when examining various formulations of discriminative learning. In the future, examining various choices of restoration tasks and searching for optimal collaborative learning strategies may lead to even stronger representations for medical imaging. In this paper, we have focused on medical imaging, but we envision that DiRA can also offer outstanding performance for vision tasks that demand fine-grained details. Acknowledgments: With the help of Zongwei Zhou, Zuwei Guo started implementing the earlier ideas behind “United & Unified”, which has branched out into DiRA. We thank them for their feasibility exploration, especially their initial evaluation on TransVW [26] and various training strategies. This research has been supported in part by ASU and Mayo Clinic through a Seed Grant and an Innovation Grant and in part by the NIH under Award Number R01HL128785. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. This work utilized the GPUs provided in part by the ASU Research Computing and in part by the Extreme Science and Engineering Discovery Environment (XSEDE) funded by the National Science Foundation (NSF) under grant number ACI-1548562. Paper content is covered by patents pending.
Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Discriminative learning, restorative learning, and adversarial learning have proven beneficial for self-supervised learning schemes in computer vision and medical imaging. Existing efforts, however, omit their synergistic effects on each other in a ternary setup, which, we envision, can sig-nificantly benefit deep semantic representation learning. To realize this vision, we have developed DiRA, thefirstframework that unites discriminative, restorative, and adversarial learning in a unified manner to collaboratively glean complementary visual information from unlabeled medical images for fine-grained semantic representation learning. Our extensive experiments demonstrate that DiRA (1) encourages collaborative learning among three learning ingredients, resulting in more generalizable representation across organs, diseases, and modalities; (2) outperforms fully supervised ImageNet models and increases robustness in small data regimes, reducing annotation cost across multiple medical imaging applications; (3) learns fine-grained semantic representation, facilitating accurate lesion localization with only image-level annotation; and (4) enhances state-of-the-art restorative approaches, revealing that DiRA is a general mechanism for united representation learning. All code and pretrained models are available at https://github.com/JLiangLab/DiRA.
AB - Discriminative learning, restorative learning, and adversarial learning have proven beneficial for self-supervised learning schemes in computer vision and medical imaging. Existing efforts, however, omit their synergistic effects on each other in a ternary setup, which, we envision, can sig-nificantly benefit deep semantic representation learning. To realize this vision, we have developed DiRA, thefirstframework that unites discriminative, restorative, and adversarial learning in a unified manner to collaboratively glean complementary visual information from unlabeled medical images for fine-grained semantic representation learning. Our extensive experiments demonstrate that DiRA (1) encourages collaborative learning among three learning ingredients, resulting in more generalizable representation across organs, diseases, and modalities; (2) outperforms fully supervised ImageNet models and increases robustness in small data regimes, reducing annotation cost across multiple medical imaging applications; (3) learns fine-grained semantic representation, facilitating accurate lesion localization with only image-level annotation; and (4) enhances state-of-the-art restorative approaches, revealing that DiRA is a general mechanism for united representation learning. All code and pretrained models are available at https://github.com/JLiangLab/DiRA.
KW - biological and cell microscopy
KW - Medical
KW - Self-& semi-& meta- & unsupervised learning
UR - http://www.scopus.com/inward/record.url?scp=85129558239&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85129558239&partnerID=8YFLogxK
U2 - 10.1109/CVPR52688.2022.02016
DO - 10.1109/CVPR52688.2022.02016
M3 - Conference contribution
AN - SCOPUS:85129558239
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 20792
EP - 20802
BT - Proceedings - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
PB - IEEE Computer Society
Y2 - 19 June 2022 through 24 June 2022
ER -