Biology subjects

Chueh, S.

Publications and source records attributed to Chueh, S..

3 recordsLinked to original sources

bioRxiv2025

Self-Supervised Missing Wedge Correction in Soft X-Ray Tomography: Towards Accurate Cellular Morphology and Volume Quantification

Soft X-Ray tomography (SXT) is a non-invasive bio-imaging technique that enables 3D imaging of cellular structures in large volume, with a unique resolution range that bridges the gap between fluorescence and transmission electron microscopy. However, a fundamental limitation, the missing wedge artefact caused by incomplete tilt-series acquisition, introduces systematic structural elongation in the reconstructed tomograms. This artifact compromises accurate quantitative biological analysis by overestimating cellular and organelle volumes. To overcome this persistent issue, we introduce a novel, self-supervised missing wedge correction model that learns key sine-wave patterns from existing SXT sinograms of the tilt series stacks. This model can be applied to recover the missing-angle region of the sinogram, reducing distortions and elongations in the reconstructed tomograms. We demonstrate a significant quantitative improvement in artifact removal, achieving faithful recovery of the spherical morphology of lipid droplets. We further applied this method to Plasmodium falciparum hemozoin crystals, a biomarker in antimalarial drug efficacy studies. Our model successfully reduced volume overestimation, achieving up to a 16% decrease in distorted volume. This level of precision is paramount for correctly interpreting the mode of action of antimalarial drugs.

cell biology↗

bioRxiv2025

SXTractor: A Self-Supervised Feature Extractor of Soft X-Ray Images That Enables Few-Shot Tomogram Segmentation

Soft X-ray tomography (SXT) is a powerful, non-invasive bio-imaging technique that enables visualization of cellular structures in near-native states. Despite its potential, the development of dedicated image analysis tools -- particularly deep-learning-based models -- has been limited, largely due to the limited accessibility of soft X-ray microscopes and the scarcity of labeled SXT data. To address this deficit, in this work, we present SXTractor, a self-supervised SXT feature extractor based on the DINO framework. SXTractor can be fine-tuned with minimal labeled data and effectively adapted to various downstream tasks. We demonstrate its utility on few-shot tomogram segmentation, where it significantly outperforms the model when trained from scratch. Furthermore, it achieves few-shot segmentation performance comparable to that of the Segment Anything Model (SAM), despite SAM being a segmentation-specific model pretrained on millions of labeled images with a significantly larger model size. Most importantly, SXTractor enables a diverse range of downstream applications of deep learning to SXT, thus offering a practical and scalable solution for SXT image analysis in data-constrained settings.

bioengineering↗

bioRxiv2025

Robust Deep Denoising of Soft X-Ray Tomography Data for Biological Research: Targeting Tilt Series Versus Reconstructed Tomograms

1.Soft X-ray microscopy (SXM) is a powerful tool for nanoscale 3D imaging of hydrated, intact cells. However, its application is limited by pixel-wise-correlated noise inherent in the imaging process. While the deep-learning-based framework Noise2Inverse has been proposed for tomography denoising, the lack of robust validation for newly revealed features raises concerns regarding their practical biological utility. We argue that denoising the raw tilt series, instead of on the reconstructed tomogram, offers a significant advantage by leveraging the tomographic reconstruction process to mitigate local prediction errors through averaging. Consequently, features present in the tomogram reconstructed from a denoised tilt series exhibit higher credibility, as falsely predicted details on certain tilt series slices are likely to be averaged out. Nevertheless, denoising the tilt series remains challenging due to the spatially correlated noise that occurs in bioimaging. Existing self-supervised methods relying on a single noisy image struggle with such noise, while the Noise2Noise framework necessitates paired noisy datasets for training. This study addresses these challenges by investigating practical imaging workflows in SXM and related bioimaging modalities to identify existing imaging resources as training data for tilt series denoising. We compare the denoising performance when applied to tilt series versus reconstructed tomograms and evaluate the efficacy of chosen methods on real biological specimens to assess their applicability within practical SXM research. Our findings establish a robust and efficient denoising strategy for SXM by directly addressing the complexities of noise in the raw tilt series. HighlightsO_LITargeting tilt series for denoising inherently validates the correctness of the denoised tomogram via the reconstruction process. C_LIO_LIExisting multi-frame imaging schemes provide necessary resources for training Noise2Noise framework for tilt series denoising. C_LIO_LINoise2Noise applied to tilt series reveals finer and more reliable details than direct denoising on reconstructed tomograms (Noise2Inverse). C_LIO_LIDenoising tilt series minimizes inference time compared to Noise2Inverse in soft X-ray tomography. C_LI

bioengineering↗