Iterative-in-Iterative Super-Resolution Biomedical Imaging Using One Real Image

TL;DR

This paper introduces a novel iterative training method for biomedical image super-resolution that uses only a single real image, leveraging self-generated data to improve image quality without extensive datasets.

Contribution

The study presents a self-revolutionary training approach for deep learning super-resolution models using only one real image, eliminating the need for large datasets.

Findings

7.5% improvement in structural similarity

5.49% increase in peak-signal-to-noise ratio

Consistent visual enhancement of biomedical images

Abstract

Deep learning-based super-resolution models have the potential to revolutionize biomedical imaging and diagnoses by effectively tackling various challenges associated with early detection, personalized medicine, and clinical automation. However, the requirement of an extensive collection of high-resolution images presents limitations for widespread adoption in clinical practice. In our experiment, we proposed an approach to effectively train the deep learning-based super-resolution models using only one real image by leveraging self-generated high-resolution images. We employed a mixed metric of image screening to automatically select images with a distribution similar to ground truth, creating an incrementally curated training data set that encourages the model to generate improved images over time. After five training iterations, the proposed deep learning-based super-resolution model experienced a 7.5\% and 5.49\% improvement in structural similarity and peak-signal-to-noise ratio, respectively. Significantly, the model consistently produces visually enhanced results for training, improving its performance while preserving the characteristics of original biomedical images. These findings indicate a potential way to train a deep neural network in a self-revolution manner independent of real-world human data.