Learning the Effect of Registration Hyperparameters with HyperMorph
arXiv:2203.16680 · doi:10.59275/j.melba.2022-74f1
Abstract
We introduce HyperMorph, a framework that facilitates efficient hyperparameter tuning in learning-based deformable image registration. Classical registration algorithms perform an iterative pair-wise optimization to compute a deformation field that aligns two images. Recent learning-based approaches leverage large image datasets to learn a function that rapidly estimates a deformation for a given image pair. In both strategies, the accuracy of the resulting spatial correspondences is strongly influenced by the choice of certain hyperparameter values. However, an effective hyperparameter search consumes substantial time and human effort as it often involves training multiple models for different fixed hyperparameter values and may lead to suboptimal registration. We propose an amortized hyperparameter learning strategy to alleviate this burden by learning the impact of hyperparameters on deformation fields. We design a meta network, or hypernetwork, that predicts the parameters of a registration network for input hyperparameters, thereby comprising a single model that generates the optimal deformation field corresponding to given hyperparameter values. This strategy enables fast, high-resolution hyperparameter search at test-time, reducing the inefficiency of traditional approaches while increasing flexibility. We also demonstrate additional benefits of HyperMorph, including enhanced robustness to model initialization and the ability to rapidly identify optimal hyperparameter values specific to a dataset, image contrast, task, or even anatomical region, all without the need to retrain models. We make our code publicly available at http://hypermorph.voxelmorph.net.
arXiv commentAccepted for publication at the Journal of Machine Learning for Biomedical Imaging (MELBA) at https://www.melba-journal.org
References in corpus (30)
- Adam: A Method for Stochastic Optimization
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
- Practical Bayesian Optimization of Machine Learning Algorithms
- VoxelMorph: A Learning Framework for Deformable Medical Image Registration
- Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimization
- A Deep Learning Framework for Unsupervised Affine and Deformable Image Registration
- A Learned Representation For Artistic Style
- Weakly-Supervised Convolutional Neural Networks for Multimodal Image Registration
- Unsupervised Learning of Probabilistic Diffeomorphic Registration for Images and Surfaces
- Gradient-based Hyperparameter Optimization through Reversible Learning
- SMASH: One-Shot Model Architecture Search through HyperNetworks
- Recursive Cascaded Networks for Unsupervised Medical Image Registration
- Learning a Probabilistic Model for Diffeomorphic Registration
- Non-stochastic Best Arm Identification and Hyperparameter Optimization
- A Probabilistic U-Net for Segmentation of Ambiguous Images
- Hyperparameter optimization with approximate gradient
- MetaPruning: Meta Learning for Automatic Neural Network Channel Pruning
- Multi-fidelity Bayesian Optimisation with Continuous Approximations
- HyperNetworks
- Scalable Gradient-Based Tuning of Continuous Regularization Hyperparameters
- Stochastic Hyperparameter Optimization through Hypernetworks
- Learning Conditional Deformable Templates with Convolutional Networks
- Hypernetwork functional image representation
- Forward and Reverse Gradient-Based Hyperparameter Optimization
- Principled Weight Initialization for Hypernetworks
- Fast GPU 3D Diffeomorphic Image Registration
- HyperGAN: A Generative Model for Diverse, Performant Neural Networks
- DHP: Differentiable Meta Pruning via HyperNetworks
- Modular Universal Reparameterization: Deep Multi-task Learning Across Diverse Domains