arXiv subjects

Hélène Bouvrais

Publications and source records attributed to Hélène Bouvrais.

6 recordsLinked to original sources

arXiv2026

Flow Matching Meets 3D Curvilinear Structure Segmentation in Medical Imaging

Segmentation of curvilinear anatomical structures in 3D medical images remains challenging due to complex topology, severe class imbalance, weak contrast, and large variations in structure morphology. While deep learning approaches for 3D curvilinear segmentation have been proposed, they are often tailored to specific anatomies or modalities, limiting generalization across clinical settings and leaving room for improvement. Recent generative models have shown the benefits of iterative prediction for structured segmentation tasks, yet diffusion-based methods suffer from computationally expensive sampling, hindering their use on high-resolution 3D volumes. We present 3D-CurvSegFlow, a flow matching-based model for 3D curvilinear structure segmentation. The model learns a continuous transformation from a simple source distribution to the target vascular representation, enabling progressive refinement of complex curvilinear geometries with efficient inference. We evaluate our method on Three public challenging datasets covering distinct anatomies and modalities: portal vein, cerebral vessel, and coronary arteries. Using a common architecture and training strategy across all tasks, our method outperforms general-purpose and vessel-specific approaches, with strong preservation of thin branches and vascular continuity. This work not only advances the state-of-the-art in 3D curvilinear segmentation but also opens new avenues for efficient, generalizable, and clinically applicable methods in medical image analysis.

cs.CV↗

arXiv2026

MTCurv: Deep learning for direct microtubule curvature mapping in noisy fluorescence microscopy images

Accurate quantification of the geometry of curvilinear biological structures is essential for understanding cellular mechanics and disease-related morphological alterations. Microtubule curvature is a key descriptor of filament rigidity and mechanical perturbations. However, reliable curvature extraction from fluorescence microscopy images remains challenging due to noise, low contrast, and partial filament visibility. Existing approaches rely on segmentation pipelines with pre or post-processing, which are highly sensitive to segmentation errors and often fail under adverse imaging conditions. In this work, we propose MTCurv, a deep learning framework for direct, segmenta-tion-free regression of microtubule curvature maps from noisy microscopy images. Leveraging a synthetic dataset with pixel-wise curvature annotations, we reformulated curvature estimation as a regression problem and adapted an attention-based residual U-Net. To reduce hallucinations and enforce spatial coherence, we introduced a gradient-aware loss combining Mean Squared Error with a gradient consistency term. Beyond model and loss design, we evaluated commonly used regression and image quality metrics, revealing that many perceptual and blind metrics are poorly suited for curvature estimation. Correlation-based metrics, particularly Spearman correlation, emerged as more reliable indicators of curvature prediction quality. Experiments on two datasets of increasing difficulty demonstrated that MTCurv accurately recovers local microtubule curvatures, even in the presence of background fluorescence. Ablation studies highlighted the contribution of both residual encoding and attention-based decoding. Overall, this work provides a practical tool for filament curvature analysis and methodological insights for geometry-aware regression in biomedical imaging. Datasets and code are made available.

cs.CV↗

arXiv2026

A novel attention mechanism for noise-adaptive and robust segmentation of microtubules in microscopy images

Segmenting cytoskeletal filaments in microscopy images is essential for studying their roles in cellular processes. However, this task is highly challenging due to the fine, densely packed, and intertwined nature of these structures. Imaging limitations further complicate analysis. While deep learning has advanced segmentation of large, well-defined biological structures, its performance often degrades under such adverse conditions. Additional challenges include obtaining precise annotations for curvilinear structures and managing severe class imbalance during training. We introduce a novel noise-adaptive attention mechanism that extends the Squeeze-and-Excitation (SE) module to dynamically adjust to varying noise levels. Integrated into a U-Net decoder with residual encoder blocks, this yields ASE_Res_UNet, a lightweight yet high-performance model. We also developed a synthetic dataset generation strategy that ensures accurate annotations of fine filaments in noisy images. We systematically evaluated loss functions and metrics to mitigate class imbalance, ensuring robust performance assessment. ASE_Res_UNet effectively segmented microtubules in noisy synthetic images, outperforming its ablated variants. It also demonstrated superior segmentation compared to models with alternative attention mechanisms or distinct architectures, while requiring fewer parameters, making it efficient for resource-constrained environments. Evaluation on a newly curated real microscopy dataset and a recently reannotated dataset highlighted ASE_Res_UNet's effectiveness in segmenting microtubules beyond synthetic images. For these datasets, ASE_Res_UNet was competitive with a recent synthetic data-driven approach that shares two cytoskeleton pretrained models. Importantly, ASE_Res_UNet showed strong transferability to other curvilinear structures (blood vessels and nerves) across diverse imaging conditions.

q-bio.QM↗

arXiv2026

MTFlow: Time-Conditioned Flow Matching for Microtubule Segmentation in Noisy Microscopy Images

Microtubules are cytoskeletal filaments that play essential roles in many cellular processes and are key therapeutic targets in several diseases. Accurate segmentation of microtubule networks is critical for studying their organization and dynamics but remains challenging due to filament curvature, dense crossings, and image noise. We present MTFlow, a novel time-conditioned flow-matching model for microtubule segmentation. Unlike conventional U-Net variants that predict masks in a single pass, MTFlow learns vector fields that iteratively transport noisy masks toward the ground truth, enabling interpretable, trajectory-based refinement. Our architecture combines a U-Net backbone with temporal embeddings, allowing the model to capture the dynamics of uncertainty resolution along filament boundaries. We trained and evaluated MTFlow on synthetic and real microtubule datasets and assessed its generalization capability on public biomedical datasets of curvilinear structures such as retinal blood vessels and nerves. MTFlow achieves competitive segmentation accuracy comparable to state-of-the-art models, offering a powerful and time-efficient tool for filamentous structure analysis with more precise annotations than manual or semi-automatic approaches.

cs.CV↗

arXiv2025

Kinesin-12 KLP-18 contributes to the kinetochore-microtubule poleward flux during the metaphase of C. elegans one-cell embryo

The mitotic spindle partitions chromosomes during cell division by connecting the poles to kinetochores through microtubules (MTs). Their plus-ends, facing the chromosomes, exhibit dynamic instability, which is critical for proper attachment. The poleward flux implicates the displacement of Mts towards the spindle poles, while plus-ends polymerise. It may result from minus-end depolymerisation (treadmilling), sliding by kinesins (e.g., Kinesin-5), or pushing by chromokinesins. Intriguingly, such flux had not been reported in the C. elegans zygote, despite homologs of flux-associated proteins being present. To investigate this, we fluorescently labelled Mts and used photobleaching. We observed no global flux; instead, the bleached zone's edges moved inward. The centrosome-facing front reflected MT dynamic instability, but the chromosome-facing front showed faster recovery, suggesting an additional mechanism. This extra velocity was spatially restricted to the vicinity of chromosomes, suggesting that only the kinetochore Mts may undergo flux. Supporting this, flux required key kinetochore regulators: NDC-80, $\text{CLS-2}^\text{CLASP}$, and $\text{ZYG-9}^\text{XMAP215}$. Flux declined as metaphase progressed, correlating with the attachment maturation from lateral to end-on, and was suppressed by SKA-1 recruitment. Classic treadmilling was unlikely, as most kinetochore MTs in C. $elegans$ do not reach spindle poles. Instead, depleting $\text{KLP-18}^\text{KIF15}$, a kinesin that cross-links and organises Mts during meiosis, reduced front movement. We propose that only kinetochore Mts undergo flux, sliding along the spindle Mts, likely powered by KLP-18. This localised sliding contrasts with global flux seen in other systems, and aligns with observations in human cells showing flux reduction as chromosome-to-pole distance increases.

q-bio.SC↗

arXiv2018

LET-99-dependent spatial restriction of active force generators makes spindle's position robust

During the asymmetric division of the Caenorhabditis elegans nematode zygote, the polarity cues distribution and daughter cell fates depend on the correct positioning of the mitotic spindle, which results from both centering and cortical pulling forces. Revealed by anaphase spindle rocking, these pulling forces are regulated by the force generator dynamics, which are in turn consequent of mitotic progression. We found a novel, additional, regulation of these forces by the spindle position. It controls astral microtubule availability at the cortex, on which the active force generators can pull. Importantly, this positional control relies on the polarity dependent LET-99 cortical band, which restricts or concentrates generators to a posterior crescent. We ascribed this control to the microtubule dynamics at the cortex. Indeed, in mapping the cortical contacts, we found a correlation between the centrosome-cortex distance and the microtubule contact density. In turn, it modulates pulling force generator activity. We modelled this control, predicting and experimentally validating that the posterior crescent extent controlled where the anaphase oscillations started, in addition to mitotic progression. Finally, we propose that spatially restricting force generator to a posterior crescent sets the spindle's final position, reflecting polarity through the LET-99 dependent restriction of force generators to a posterior crescent. This regulation superimposes that of force generator processivity. This novel control confers a low dependence on microtubule and active force generator exact numbers or dynamics, provided that they exceed the threshold needed for posterior displacement. Interestingly, this robustness originates in cell mechanics rather than biochemical networks.

q-bio.SC↗