Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Domain Generalization for Medical Imaging Classification with Linear-Dependency Regularization

About

Recently, we have witnessed great progress in the field of medical imaging classification by adopting deep neural networks. However, the recent advanced models still require accessing sufficiently large and representative datasets for training, which is often unfeasible in clinically realistic environments. When trained on limited datasets, the deep neural network is lack of generalization capability, as the trained deep neural network on data within a certain distribution (e.g. the data captured by a certain device vendor or patient population) may not be able to generalize to the data with another distribution. In this paper, we introduce a simple but effective approach to improve the generalization capability of deep neural networks in the field of medical imaging classification. Motivated by the observation that the domain variability of the medical images is to some extent compact, we propose to learn a representative feature space through variational encoding with a novel linear-dependency regularization term to capture the shareable information among medical data collected from different domains. As a result, the trained neural network is expected to equip with better generalization capability to the "unseen" medical data. Experimental results on two challenging medical imaging classification tasks indicate that our method can achieve better cross-domain generalization capability compared with state-of-the-art baselines.

Haoliang Li, YuFei Wang, Renjie Wan, Shiqi Wang, Tie-Qiang Li, Alex C. Kot• 2020

Related benchmarks

TaskDatasetResultRank
Cardiac Image SegmentationM&Ms Target A 1.0
Dice (%)82.62
15
Cardiac Image SegmentationM&Ms Target C 1.0
Dice Score0.8649
15
Cardiac Image SegmentationM&Ms (Target C)
Hausdorff Distance13.52
15
Cardiac Image SegmentationM&Ms Target B 1.0
Dice Score85.68
15
Cardiac Image SegmentationM&Ms Target D 1.0
Dice Score86.73
15
Spinal Cord Gray Matter SegmentationSCGM Site 4
Hausdorff Distance (HD)2.12
10
Spinal Cord Gray Matter SegmentationSCGM (test)
Target 1 Score88.21
10
Cardiac Image SegmentationM&Ms Target A v1
Hausdorff Distance23.35
10
Cardiac Image SegmentationM&Ms Target D
Hausdorff Distance15.8
10
Cardiac Image SegmentationM&Ms Average
Hausdorff Distance19.21
10
Showing 10 of 22 rows

Other info

Follow for update