TY - GEN
T1 - MetaMedSeg
T2 - 4th MICCAI Workshop on Domain Adaptation and Representation Transfer, DART 2022, held in conjunction with the 25th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2022
AU - Farshad, Azade
AU - Makarevich, Anastasia
AU - Belagiannis, Vasileios
AU - Navab, Nassir
N1 - Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2022
Y1 - 2022
N2 - The lack of sufficient annotated image data is a common issue in medical image segmentation. For some organs and densities, the annotation may be scarce, leading to poor model training convergence, while other organs have plenty of annotated data. In this work, we present MetaMedSeg, a gradient-based meta-learning algorithm that redefines the meta-learning task for the volumetric medical data with the goal of capturing the variety between the slices. We also explore different weighting schemes for gradients aggregation, arguing that different tasks might have different complexity and hence, contribute differently to the initialization. We propose an importance-aware weighting scheme to train our model. In the experiments, we evaluate our method on the medical decathlon dataset by extracting 2D slices from CT and MRI volumes of different organs and performing semantic segmentation. The results show that our proposed volumetric task definition leads to up to 30 % improvement in terms of IoU compared to related baselines. The proposed update rule is also shown to improve the performance for complex scenarios where the data distribution of the target organ is very different from the source organs. (Project page: http://metamedseg.github.io/
AB - The lack of sufficient annotated image data is a common issue in medical image segmentation. For some organs and densities, the annotation may be scarce, leading to poor model training convergence, while other organs have plenty of annotated data. In this work, we present MetaMedSeg, a gradient-based meta-learning algorithm that redefines the meta-learning task for the volumetric medical data with the goal of capturing the variety between the slices. We also explore different weighting schemes for gradients aggregation, arguing that different tasks might have different complexity and hence, contribute differently to the initialization. We propose an importance-aware weighting scheme to train our model. In the experiments, we evaluate our method on the medical decathlon dataset by extracting 2D slices from CT and MRI volumes of different organs and performing semantic segmentation. The results show that our proposed volumetric task definition leads to up to 30 % improvement in terms of IoU compared to related baselines. The proposed update rule is also shown to improve the performance for complex scenarios where the data distribution of the target organ is very different from the source organs. (Project page: http://metamedseg.github.io/
UR - http://www.scopus.com/inward/record.url?scp=85140452198&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-16852-9_5
DO - 10.1007/978-3-031-16852-9_5
M3 - Conference contribution
AN - SCOPUS:85140452198
SN - 9783031168512
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 45
EP - 55
BT - Domain Adaptation and Representation Transfer - 4th MICCAI Workshop, DART 2022, Held in Conjunction with MICCAI 2022, Proceedings
A2 - Kamnitsas, Konstantinos
A2 - Koch, Lisa
A2 - Islam, Mobarakol
A2 - Xu, Ziyue
A2 - Cardoso, Jorge
A2 - Dou, Qi
A2 - Rieke, Nicola
A2 - Tsaftaris, Sotirios
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 22 September 2022 through 22 September 2022
ER -