NCF: Neural Correspondence Field for Medical Image Registration
This addresses the problem of medical image registration for clinicians or researchers by enabling learning from just one data pair, though it appears incremental as it builds on existing learning-based methods.
The paper tackled the challenge of generalizable deformable image registration in medical imaging with scarce data by proposing a training-data-free learning-based method, Neural Correspondence Field (NCF), which achieved superior performance on a Lung CT dataset and outperformed a traditional method on a head and neck dataset.
Deformable image registration is a fundamental task in medical image processing. Traditional optimization-based methods often struggle with accuracy in dealing with complex deformation. Recently, learning-based methods have achieved good performance on public datasets, but the scarcity of medical image data makes it challenging to build a generalizable model to handle diverse real-world scenarios. To address this, we propose a training-data-free learning-based method, Neural Correspondence Field (NCF), which can learn from just one data pair. Our approach employs a compact neural network to model the correspondence field and optimize model parameters for each individual image pair. Consequently, each pair has a unique set of network weights. Notably, our model is highly efficient, utilizing only 0.06 million parameters. Evaluation results showed that the proposed method achieved superior performance on a public Lung CT dataset and outperformed a traditional method on a head and neck dataset, demonstrating both its effectiveness and efficiency.