Predicting Scores of Medical Imaging Segmentation Methods with Meta-Learning
This addresses the challenge of model selection for researchers and practitioners in medical imaging, though it is incremental as it applies existing meta-learning methods to this domain.
The paper tackles the problem of efficiently choosing deep learning models for medical image segmentation by using meta-learning to predict model performance on new tasks, achieving Dice scores within 0.10 of true performance on external test datasets.
Deep learning has led to state-of-the-art results for many medical imaging tasks, such as segmentation of different anatomical structures. With the increased numbers of deep learning publications and openly available code, the approach to choosing a model for a new task becomes more complicated, while time and (computational) resources are limited. A possible solution to choosing a model efficiently is meta-learning, a learning method in which prior performance of a model is used to predict the performance for new tasks. We investigate meta-learning for segmentation across ten datasets of different organs and modalities. We propose four ways to represent each dataset by meta-features: one based on statistical features of the images and three are based on deep learning features. We use support vector regression and deep neural networks to learn the relationship between the meta-features and prior model performance. On three external test datasets these methods give Dice scores within 0.10 of the true performance. These results demonstrate the potential of meta-learning in medical imaging.