Unsupervised Learning for Inverse Problems in Computed Tomography

TL;DR

This paper introduces an unsupervised deep learning method for CT image reconstruction that does not require ground-truth data, achieving high-quality results and faster processing suitable for real-time medical imaging.

Contribution

The study presents a novel unsupervised deep learning approach leveraging neural network inference for CT reconstruction, eliminating the need for supervised training with ground-truth images.

Findings

Outperforms traditional FBP and ML methods in MSE and SSIM

Achieves comparable results to supervised deep learning methods

Reduces reconstruction time significantly

Abstract

Assume you encounter an inverse problem that shall be solved for a large number of data, but no ground-truth data is available. To emulate this encounter, in this study, we assume it is unknown how to solve the imaging problem of Computed Tomography (CT). An unsupervised deep learning approach is introduced, that leverages the inherent similarities between deep neural network training, deep image prior (DIP) and unrolled optimization schemes. We demonstrate the feasibility of reconstructing images from measurement data by pure network inference, without relying on ground-truth images in the training process or additional gradient steps for unseen samples. Our method is evaluated on the two-dimensional 2DeteCT dataset, showcasing superior performance in terms of mean squared error (MSE) and structural similarity index (SSIM) compared to traditional filtered backprojection (FBP) and maximum likelihood (ML) reconstruction techniques as well as similar performance compared to a supervised DL reconstruction. Additionally, our approach significantly reduces reconstruction time, making it a promising alternative for real-time medical imaging applications. Future work will focus on extending this methodology for adaptability of the projection geometry and other use-cases in medical imaging.