Domain-Aware Few-Shot Learning for Optical Coherence Tomography Noise Reduction
This addresses the challenge of adapting noise reduction models to unseen OCT domains, which is important for medical imaging applications, though it appears incremental as it builds on existing few-shot and supervised learning methods.
The paper tackles the problem of domain shift in supervised learning for optical coherence tomography (OCT) noise reduction by proposing a few-shot learning framework that requires only a single image and ground truth for training, resulting in dramatic increases in training speed and improved sample complexity and generalization.
Speckle noise has long been an extensively studied problem in medical imaging. In recent years, there have been significant advances in leveraging deep learning methods for noise reduction. Nevertheless, adaptation of supervised learning models to unseen domains remains a challenging problem. Specifically, deep neural networks (DNNs) trained for computational imaging tasks are vulnerable to changes in the acquisition system's physical parameters, such as: sampling space, resolution, and contrast. Even within the same acquisition system, performance degrades across datasets of different biological tissues. In this work, we propose a few-shot supervised learning framework for optical coherence tomography (OCT) noise reduction, that offers a dramatic increase in training speed and requires only a single image, or part of an image, and a corresponding speckle suppressed ground truth, for training. Furthermore, we formulate the domain shift problem for OCT diverse imaging systems, and prove that the output resolution of a despeckling trained model is determined by the source domain resolution. We also provide possible remedies. We propose different practical implementations of our approach, verify and compare their applicability, robustness, and computational efficiency. Our results demonstrate significant potential for generally improving sample complexity, generalization, and time efficiency, for coherent and non-coherent noise reduction via supervised learning models, that can also be leveraged for other real-time computer vision applications.