Gaussian Process Emulators for Few-Shot Segmentation in Cardiac MRI
This work addresses the need for reduced data dependency in cardiac MRI segmentation, which is crucial for diagnosing cardiovascular diseases, but it is incremental as it builds on existing few-shot and U-Net approaches.
The paper tackles the problem of segmenting cardiac MRI with limited labeled data by introducing a method that combines few-shot learning, U-Net, and Gaussian Process Emulators, achieving higher DICE coefficients than state-of-the-art methods, especially with small support sets.
Segmentation of cardiac magnetic resonance images (MRI) is crucial for the analysis and assessment of cardiac function, helping to diagnose and treat various cardiovascular diseases. Most recent techniques rely on deep learning and usually require an extensive amount of labeled data. To overcome this problem, few-shot learning has the capability of reducing data dependency on labeled data. In this work, we introduce a new method that merges few-shot learning with a U-Net architecture and Gaussian Process Emulators (GPEs), enhancing data integration from a support set for improved performance. GPEs are trained to learn the relation between the support images and the corresponding masks in latent space, facilitating the segmentation of unseen query images given only a small labeled support set at inference. We test our model with the M&Ms-2 public dataset to assess its ability to segment the heart in cardiac magnetic resonance imaging from different orientations, and compare it with state-of-the-art unsupervised and few-shot methods. Our architecture shows higher DICE coefficients compared to these methods, especially in the more challenging setups where the size of the support set is considerably small.