Saved in:
Bibliographic Details
Main Authors: Laydi, Achraf Ait, Moctar, Sidi Mohamed Sid'El, Mourabit, Yousef El, Bouvrais, Hélène
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.26517
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914516636270592
author Laydi, Achraf Ait
Moctar, Sidi Mohamed Sid'El
Mourabit, Yousef El
Bouvrais, Hélène
author_facet Laydi, Achraf Ait
Moctar, Sidi Mohamed Sid'El
Mourabit, Yousef El
Bouvrais, Hélène
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2604_26517
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MTCurv: Deep learning for direct microtubule curvature mapping in noisy fluorescence microscopy images
Laydi, Achraf Ait
Moctar, Sidi Mohamed Sid'El
Mourabit, Yousef El
Bouvrais, Hélène
Computer Vision and Pattern Recognition
Cell Behavior
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.
title MTCurv: Deep learning for direct microtubule curvature mapping in noisy fluorescence microscopy images
topic Computer Vision and Pattern Recognition
Cell Behavior
url https://arxiv.org/abs/2604.26517