Pixel‑Dependent Noise Variance Stabilization for Learning‑Based Denoising in Imaging Systems with Radially Symmetric Beams

Pablo Corral Arroyo, Fasil Gadjimuradov, Sai Gokul Hariharan

Keywords: Image denoising, UNETs, Convolutional Neural Networks

TL;DR: Pixel-dependent Noise Varaince Stabilization to improve deep-learning based denoising of x-ray systems

Registration Requirement: Yes

Abstract: X-ray guided medical procedures may expose the patient to a non-negligible amount of radiation dose. To mitigate the dose and reduce the risk of potentially correlated health issues, it is important to optimize the radiation exposure for both the patients and clinical staff. This means that the applied radiation dose should be as low as reasonably achievable while ensuring that the required image quality is reached. UNETs have become the state-of-the-art denoising algorithms. In this article we show a preprocessing algorithm, which improves the robustness and generalization of denosing models of systems whose images have position-dependent x-ray intensity.

Visa & Travel: Yes

Read CFP & Author Instructions: Yes

Originality Policy: Yes

Single-blind & Not Under Review Elsewhere: Yes

LLM Policy: Yes

Submission Number: 104