Image quality assessment for machine learning tasks using meta-reinforcement learning

TL;DR

This paper introduces a meta-reinforcement learning framework for image quality assessment tailored to specific machine learning tasks, enabling more accurate and adaptable evaluation of images for clinical applications.

Contribution

It proposes a novel task-specific, adaptable IQA method using meta-reinforcement learning that can be trained jointly with task predictors for improved performance.

Findings

Effective in clinical ultrasound-guided prostate intervention

Improves pneumonia detection accuracy on X-ray images

Enhances adaptability of IQA to new datasets

Abstract

In this paper, we consider image quality assessment (IQA) as a measure of how images are amenable with respect to a given downstream task, or task amenability. When the task is performed using machine learning algorithms, such as a neural-network-based task predictor for image classification or segmentation, the performance of the task predictor provides an objective estimate of task amenability. In this work, we use an IQA controller to predict the task amenability which, itself being parameterised by neural networks, can be trained simultaneously with the task predictor. We further develop a meta-reinforcement learning framework to improve the adaptability for both IQA controllers and task predictors, such that they can be fine-tuned efficiently on new datasets or meta-tasks. We demonstrate the efficacy of the proposed task-specific, adaptable IQA approach, using two clinical applications for ultrasound-guided prostate intervention and pneumonia detection on X-ray images.