Adversarial training with cycle consistency for unsupervised super-resolution in endomicroscopy

Abstract

In recent years, endomicroscopy has become increasingly used for diagnostic purposes and interventional guidance. It can provide intraoperative aids for real-time tissue characterization and can help to perform visual investigations aimed for example to discover epithelial cancers. Due to physical constraints on the acquisition process, endomicroscopy images, still today have a low number of informative pixels which hampers their quality. Post-processing techniques, such as Super-Resolution (SR), are a potential solution to increase the quality of these images. SR techniques are often supervised, requiring aligned pairs of low-resolution (LR) and high-resolution (HR) images patches to train a model. However, in our domain, the lack of HR images hinders the collection of such pairs and makes supervised training unsuitable. For this reason, we propose an unsupervised SR framework based on an adversarial deep neural network with a physically-inspired cycle consistency, designed to impose some acquisition properties on the super-resolved images. Our framework can exploit HR images, regardless of the domain where they are coming from, to transfer the quality of the HR images to the initial LR images. This property can be particularly useful in all situations where pairs of LR/HR are not available during the training. Our quantitative analysis, validated using a database of 238 endomicroscopy video sequences from 143 patients, shows the ability of the pipeline to produce convincing super-resolved images. A Mean Opinion Score (MOS) study also confirms this quantitative image quality assessment.

Accepted for publication on Medical Image Analysis journal

References in corpus (2)

• Adversarial Variational Bayes: Unifying Variational Autoencoders and Generative Adversarial Networks
• Effective deep learning training for single-image super-resolution in endomicroscopy exploiting video-registration-based reconstruction

Cited by in corpus (14)

• Generative Adversarial Network in Medical Imaging: A Review
• High-throughput molecular imaging via deep learning enabled Raman spectroscopy
• CAI4CAI: The Rise of Contextual Artificial Intelligence in Computer Assisted Interventions
• Deep learning in biomedical optics
• Low-field magnetic resonance image enhancement via stochastic image quality transfer
• Physics-informed GANs for Coastal Flood Visualization
• Uncertainty Quantification in Deep Learning for Safer Neuroimage Enhancement
• Zero-shot super-resolution with a physically-motivated downsampling kernel for endomicroscopy
• Learning from Irregularly Sampled Data for Endomicroscopy Super-resolution: A Comparative Study of Sparse and Dense Approaches
• A Comprehensive Review of Deep Learning-based Single Image Super-resolution
• Semi-Supervised Learning of Mutually Accelerated MRI Synthesis without Fully-Sampled Ground Truths
• Unsupervised Super-Resolution of Satellite Imagery for High Fidelity Material Label Transfer
• Multi-modality super-resolution loss for GAN-based super-resolution of clinical CT images using micro CT image database
• Micro CT Image-Assisted Cross Modality Super-Resolution of Clinical CT Images Utilizing Synthesized Training Dataset