"One-Shot" Reduction of Additive Artifacts in Medical Images
This addresses the challenge of artifact reduction in medical imaging for clinical adoption, offering a flexible solution without needing large pre-trained datasets, though it is incremental as it builds on deep learning but adapts it to one-shot scenarios.
The paper tackles the problem of reducing additive artifacts in medical images, which vary widely and limit existing deep-learning methods that rely on specific training sets, by introducing OSAR, a one-shot method that trains a lightweight network at test-time using synthesized data from the input image, achieving better artifact reduction than state-of-the-art methods in CT and MRI with shorter test times.
Medical images may contain various types of artifacts with different patterns and mixtures, which depend on many factors such as scan setting, machine condition, patients' characteristics, surrounding environment, etc. However, existing deep-learning-based artifact reduction methods are restricted by their training set with specific predetermined artifact types and patterns. As such, they have limited clinical adoption. In this paper, we introduce One-Shot medical image Artifact Reduction (OSAR), which exploits the power of deep learning but without using pre-trained general networks. Specifically, we train a light-weight image-specific artifact reduction network using data synthesized from the input image at test-time. Without requiring any prior large training data set, OSAR can work with almost any medical images that contain varying additive artifacts which are not in any existing data sets. In addition, Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) are used as vehicles and show that the proposed method can reduce artifacts better than state-of-the-art both qualitatively and quantitatively using shorter test time.