J-Invariant Volume Shuffle for Self-Supervised Cryo-Electron Tomogram Denoising on Single Noisy Volume

TL;DR

This paper introduces a novel self-supervised denoising method for Cryo-ET volumes using a J-invariant network with volume unshuffle/shuffle, achieving superior noise reduction and structural preservation from a single noisy volume.

Contribution

The proposed method is the first to utilize a J-invariant network with volume unshuffle/shuffle for effective self-supervised denoising of single noisy Cryo-ET volumes.

Findings

Outperforms existing denoising methods in noise reduction

Preserves structural details better than prior approaches

Effective with only a single noisy volume as input

Abstract

Cryo-Electron Tomography (Cryo-ET) enables detailed 3D visualization of cellular structures in near-native states but suffers from low signal-to-noise ratio due to imaging constraints. Traditional denoising methods and supervised learning approaches often struggle with complex noise patterns and the lack of paired datasets. Self-supervised methods, which utilize noisy input itself as a target, have been studied; however, existing Cryo-ET self-supervised denoising methods face significant challenges due to losing information during training and the learned incomplete noise patterns. In this paper, we propose a novel self-supervised learning model that denoises Cryo-ET volumetric images using a single noisy volume. Our method features a U-shape J-invariant blind spot network with sparse centrally masked convolutions, dilated channel attention blocks, and volume unshuffle/shuffle technique. The volume-unshuffle/shuffle technique expands receptive fields and utilizes multi-scale representations, significantly improving noise reduction and structural preservation. Experimental results demonstrate that our approach achieves superior performance compared to existing methods, advancing Cryo-ET data processing for structural biology research