publications | Daniel Franco-Barranco

2026

J Microsc
Bioimage management and analysis in Galaxy: Tools, workflows, training, and community practices

Leonid Kostrykin, Riccardo Massei, Diana Chiang, and 23 more authors

Journal of Microscopy, Sep 2026

Abs DOI Bib

Image analysis in the life sciences is constrained by fragmented software ecosystems, heterogeneous data formats, and limited reproducibility. These barriers hinder the reuse of image analysis methods and the sustainability of tools. In this article, we describe how the Galaxy platform enables FAIR (Findable, Accessible, Interoperable, and Reusable) image analysis by providing an integrated environment for data access, workflow execution, provenance tracing, and training. We present Galaxy as a computational workbench that supports diverse image formats and integrates with public, institutional, and private repositories. We describe a reference structure for FAIR image analysis workflows and illustrate how this pattern supports reproducibility, interoperability, and reuse. We also describe community-driven training and sustainability practices that embed FAIR principles directly into executable tutorials and shared workflows. Together, these foundations position Galaxy as a reproducible, scalable, and community-maintained platform for FAIR image analysis across the life sciences and beyond.
@article{kostrykin2026galaxy, title = {Bioimage management and analysis in Galaxy: Tools, workflows, training, and community practices}, author = {Kostrykin, Leonid and Massei, Riccardo and Chiang, Diana and Wollmann, Thomas and Burel, Jean-Marie and Tabernero, David Lopez and Sun, Yi and Gao, Qi and Paul, Maarten W and Videm, Pavankumar and Watson, Cameron and Franco-Barranco, Daniel and Kumar, Anup and Gou{\'e}, Nadia and Koopaei, Reyhaneh Tavakoli and Ulman, Vladim{\'i}r and Jum'ah, Khaled and Poterlowicz, Krzysztof and Etzrodt, Martin and Mu{\~n}oz-Barrutia, Arrate and Le D{\'e}v{\'e}dec, Sylvia E and Moore, Josh and Goecks, Jeremy and Gruening, Bjoern and Rohr, Karl and Serrano-Solano, Beatriz}, journal = {Journal of Microscopy}, year = {2026}, month = sep, doi = {10.1111/jmi.70165}, url = {https://doi.org/10.1111/jmi.70165}, }
Nat Aging
Three-dimensional mapping of intact ovaries reveals the aging dynamics of the ovarian reserve

Arturo D’Angelo, Daniel Franco-Barranco, Marco Musy, and 5 more authors

Nature Aging, Aug 2026

Abs DOI Bib

Female fertility depends on a finite pool of oocytes that depletes during aging, yet the spatiotemporal dynamics of this depletion remain poorly understood. Traditional methods obscure the three-dimensional architecture of the ovary, limiting quantitative insights. Here we combine light-sheet microscopy, artificial intelligence-driven segmentation and mathematical modeling to map over 85,000 oocytes in whole ovaries across the reproductive lifespan in mouse. We find that newly activated oocytes represent a fixed fraction of the total oocyte pool despite an age-related decline in oocyte numbers. Spatial analysis revealed that oocytes are enriched along the lateral ovarian axis, and local oocyte density positively correlates with activation. We also uncover a bimodal distribution of oocyte sizes, suggesting a bottleneck during oogenesis. Finally, a differential equation-based model captures the kinetics of oocyte activation and loss. Our findings establish a quantitative framework for understanding ovarian aging and suggest that an organ-scale regulatory mechanism coordinates the age-related decline in oocyte numbers.
@article{dangelo2026ovaries, title = {Three-dimensional mapping of intact ovaries reveals the aging dynamics of the ovarian reserve}, author = {D'Angelo, Arturo and Franco-Barranco, Daniel and Musy, Marco and Duran, Juan Manuel and Sharpe, James and Stroustrup, Nicholas and Arganda-Carreras, Ignacio and B{\"o}ke, Elvan}, journal = {Nature Aging}, year = {2026}, month = aug, volume = {6}, number = {8}, pages = {1580--1591}, doi = {10.1038/s43587-026-01178-z}, url = {https://doi.org/10.1038/s43587-026-01178-z}, }
Sci Rep
Automated segmentation of neurons and spinal cord structures in immunofluorescence images using SpineDL

Pablo Ruiz-Amezcua, Daniel Franco-Barranco, David Reigada, and 3 more authors

Scientific Reports, Jun 2026

Abs DOI Bib Code

In this study, we present SpineDL, an open-source deep learning (DL) approach for neuron and anatomical structure segmentation of the spinal cord in fluorescence images immunostained with NeuN and DAPI, within the context of murine models of spinal cord injury (SCI). SpineDL comprises two main modules: SpineDL-Neuron, for instance-level identification of neuronal somas; and SpineDL-Structure, for semantic segmentation of key spinal cord structures including gray matter, white matter, ependyma, and damaged tissue. To train the models, we developed the SpineDL dataset, a curated collection of 161 confocal images of mouse spinal cord, manually annotated by SCI researchers and organized into specific subsets. Both models are based on the HRNetV2-W64 architecture and were trained using state-of-the-art data augmentation and optimization techniques, implemented within the BiaPy framework. Our results demonstrate that SpineDL achieves researcher-level performance in both structural segmentation and neuron identification tasks, showing high robustness across anatomical regions and injury conditions.
@article{ruizamezcua2026spinedl, title = {Automated segmentation of neurons and spinal cord structures in immunofluorescence images using SpineDL}, author = {Ruiz-Amezcua, Pablo and Franco-Barranco, Daniel and Reigada, David and Mu{\~n}oz-Galdeano, Teresa and Maza, Rodrigo M and Nieto-Diaz, Manuel}, journal = {Scientific Reports}, year = {2026}, month = jun, volume = {16}, pages = {28309}, doi = {10.1038/s41598-026-57519-w}, url = {https://doi.org/10.1038/s41598-026-57519-w}, }
Nat Methods
Representation matters: A systematic study of instance segmentation approaches in bioimage analysis

Daniel Franco-Barranco, Samia Mohinta, Arrate Muñoz-Barrutia, and 1 more author

Nature Methods, 2026

Registered Report, accepted first revision

Abs DOI Bib Code

Accurate instance segmentation is central to quantitative bioimage analysis, enabling the study of cellular organization, morphology, and dynamics. While deep learning has driven major advances, biomedical pipelines predominantly rely on bottom-up formulations whose performance critically depends on the choice of intermediate representation. However, existing methods are typically evaluated under heterogeneous conditions, limiting fair comparison, reproducibility, and generalization. Here, we present a comprehensive and reproducible benchmark of bottom-up instance segmentation representations, encompassing binary feature maps, distance-based encodings, affinity graphs, and embedding formulations. Using six heterogeneous microscopy datasets spanning 2D and 3D modalities, we systematically assess how representation design and architectural choices affect segmentation accuracy and robustness under controlled training budgets. All experiments are implemented within the open-source BiaPy framework using standardized protocols and hyperparameter optimization. This study provides a unified and rigorous assessment of representation-driven performance in bioimage instance segmentation, supporting transparent comparison and method selection across diverse imaging applications.
@article{francobarranco2026representation, title = {Representation matters: A systematic study of instance segmentation approaches in bioimage analysis}, author = {Franco-Barranco, Daniel and Mohinta, Samia and Mu{\~n}oz-Barrutia, Arrate and Arganda-Carreras, Ignacio}, journal = {Nature Methods}, note = {Registered Report, accepted first revision}, year = {2026}, doi = {10.6084/m9.figshare.32782521.v1}, url = {https://doi.org/10.6084/m9.figshare.32782521.v1}, }
Nat Commun
SAMJ: Fast Image Annotation on ImageJ/Fiji via Segment Anything Model

Carlos Garcia-Lopez-de-Haro, Caterina Fuster-Barcelo, Curtis T. Rueden, and 9 more authors

Nature Communications, 2026

Abs DOI arXiv Bib PDF

Mask annotation remains a significant bottleneck in AI-driven biomedical image analysis due to its labor-intensive nature. To address this challenge, we introduce SAMJ, a user-friendly ImageJ/Fiji plugin leveraging the Segment Anything Model (SAM). SAMJ enables seamless, interactive annotations with one-click installation on standard computers. Designed for real-time object delineation in large scientific images, SAMJ is an easy-to-use solution that simplifies and accelerates the creation of labeled image datasets.
@article{garcia2026samj, title = {SAMJ: Fast Image Annotation on ImageJ/Fiji via Segment Anything Model}, author = {Garcia-Lopez-de-Haro, Carlos and Fuster-Barcelo, Caterina and Rueden, Curtis T. and Heras, Jonathan and Ulman, Vladimir and Franco-Barranco, Daniel and Ines, Adrian and Eliceiri, Kevin W. and Olivo-Marin, Jean-Christophe and Tinevez, Jean-Yves and Sage, Daniel and Munoz-Barrutia, Arrate}, journal = {Nature Communications}, volume = {17}, number = {1}, pages = {5402}, year = {2026}, publisher = {Nature Publishing Group}, doi = {10.1038/s41467-026-71752-x}, }

2025

bioRxiv
Beyond Agreement: Standardizing Crowdsourced Synapse Annotations through Proofreading in EM Connectomics

Shi Yan Lee, Ana Correia, Nicolo Ceffa, and 5 more authors

bioRxiv, 2025

Abs DOI Bib HTML

Reliable synapse identification in volumetric EM is hampered by subtle, 3D cues that yield variable human judgments. We present a standardized proofreading protocol that pairs explicit, operational criteria with machine-learning candidate generation and a two-stage calibration of annotators. In two larval Drosophila melanogaster volumes imaged at 8x8x8 nm, five raters (expert + 4 calibrated annotators) reviewed model-proposed candidates using efficient node-based labels. Multi-rater judgments were aggregated with a probabilistic Dawid-Skene (DS) model to produce consensus labels with calibrated uncertainty. Post-calibration, individual annotator accuracy versus the expert improved (McNemar p < 0.05 for all raters), DS-expert agreement increased, and DS posterior entropy decreased for true positives/negatives, indicating more decisive consensus; gains were modest and dataset-dependent in chance-corrected agreement (Krippendorff’s alpha). By making uncertainty explicit, this protocol converts noisy judgments into auditable supervision suitable for training and evaluation, while honestly communicating residual ambiguity essential for reliable and robust connectomics at scale.
@article{lee2025beyondagreement, title = {Beyond Agreement: Standardizing Crowdsourced Synapse Annotations through Proofreading in EM Connectomics}, author = {Lee, Shi Yan and Correia, Ana and Ceffa, Nicolo and Robbins, Miranda and Franco-Barranco, Daniel and Zlatic, Marta and Cardona, Albert and Mohinta, Samia}, journal = {bioRxiv}, year = {2025}, doi = {10.1101/2025.09.26.678851}, }
bioRxiv
Mapping the nervous system of the Idiosepius hallami pygmy squid: insights from whole-animal X-ray nanotomography imaging

Ana Correia, Wen-Sung Chung, Samia Mohinta, and 5 more authors

bioRxiv, 2025

Abs DOI Bib HTML

The study of a nervous system as big as the cephalopod’s requires multimodal imaging approaches capable of capturing neural architecture across scales. Here, we present a whole-animal volume of the pygmy squid hatchling Idiosepius hallami, acquired using X-ray holographic nanotomography at the beamline ID16A of the European Synchrotron. The reconstructed 3D volume comprises 40 tiled scans acquired at a voxel size of 125 nm. While individual neurons are not resolved at this resolution, we segmented major body regions and mapped the large-scale connectivity by tracing afferent and efferent nerve bundles, including projections from the olfactory organs, chromatophore lobes, and arm ganglia to the brain. The acquisition of this dataset represents a significant milestone for X-ray nanotomography, being the largest whole animal volume imaged at this spatial resolution. The volume serves as a resource for comparative neuroscience and cephalopod biology.
@article{correia2025idiosepius, title = {Mapping the nervous system of the Idiosepius hallami pygmy squid: insights from whole-animal X-ray nanotomography imaging}, author = {Correia, Ana and Chung, Wen-Sung and Mohinta, Samia and Franco-Barranco, Daniel and Vorobyev, Artem and Pacureanu, Alexandra and Cardona, Albert and Corrales, Marc}, journal = {bioRxiv}, year = {2025}, doi = {10.1101/2025.09.25.678516}, }
arXiv
Towards Generalized Synapse Detection Across Invertebrate Species

Samia Mohinta, Daniel Franco-Barranco, Shi Yan Lee, and 1 more author

2025

Abs arXiv Bib

Behavioural differences across organisms, whether healthy or pathological, are closely tied to the structure of their neural circuits. Yet, the fine-scale synaptic changes that give rise to these variations remain poorly understood, in part due to persistent challenges in detecting synapses reliably and at scale. Volume electron microscopy (EM) offers the resolution required to capture synaptic architecture, but automated detection remains difficult due to sparse annotations, morphological variability, and cross-dataset domain shifts. To address this, we make three key contributions. First, we curate a diverse EM benchmark spanning four datasets across two invertebrate species: adult and larval Drosophila melanogaster, and Megaphragma viggianii (micro-WASP). Second, we propose SimpSyn, a single-stage Residual U-Net trained to predict dual-channel spherical masks around pre- and post-synaptic sites, designed to prioritize training and inference speeds and annotation efficiency over architectural complexity. Third, we benchmark SimpSyn against Buhmann et al.’s Synful, a state-of-the-art multi-task model that jointly infers synaptic pairs. Despite its simplicity, SimpSyn consistently outperforms Synful in F1-score across all volumes for synaptic site detection.
@misc{mohinta2025synapse, title = {Towards Generalized Synapse Detection Across Invertebrate Species}, author = {Mohinta, Samia and Franco-Barranco, Daniel and Lee, Shi Yan and Cardona, Albert}, year = {2025}, archiveprefix = {arXiv}, url = {https://arxiv.org/abs/2509.17041}, }

Seminario médico

Integración del análisis de imagen y la secuenciación de ARN nuclear en la investigación de la lesión medular: avances recientes y perspectivas

Pablo Ruiz Amezcua, Teresa Muñoz Galdeano, David Reigada Prado, and 4 more authors

Seminario médico, 2025

Bib HTML

@article{ruizamezcua2025seminario,
  title = {Integraci{\'o}n del an{\'a}lisis de imagen y la secuenciaci{\'o}n de ARN nuclear en la investigaci{\'o}n de la lesi{\'o}n medular: avances recientes y perspectivas},
  author = {Ruiz Amezcua, Pablo and Mu{\~n}oz Galdeano, Teresa and Reigada Prado, David and Franco Barranco, Daniel and Mart{\'i}nez Maza, Rodrigo and Esteban Ruiz, Francisco J and Nieto Diaz, Manuel},
  journal = {Seminario m{\'e}dico},
  volume = {65},
  pages = {71--80},
  year = {2025},
}

ISBI
Super-Resolution Benchmarking for 3D Image-to-Image Fusion Problem

Daniel Franco-Barranco, Aitor González-Marfil, Albert Cardona, and 2 more authors

In 2025 IEEE 22nd International Symposium on Biomedical Imaging (ISBI), 2025

Abs DOI Bib PDF

Fluorescence microscopy faces challenges in resolution, phototoxicity, and anisotropic artifacts. The Fuse My Cells challenge, organized by France-BioImaging, aims to develop deep learning models that predict fused 3D volumes from single-view acquisitions, reducing phototoxic exposure while enhancing resolution. In this work, we benchmark state-of-the-art super-resolution models, including DFCAN, RCAN-3D, UNETR, and a 3D-adapted RCAN-it, evaluating their performance on the Fuse My Cells challenge dataset, which encompasses 802 3D light-sheet microscopy images. A novel training strategy prioritizing high-discrepancy regions optimizes efficiency and improves reconstruction accuracy. Our findings suggest that super-resolution models can not fully reconstruct the information on those image areas with minimum signal information.
@inproceedings{franco2025superres, title = {Super-Resolution Benchmarking for 3D Image-to-Image Fusion Problem}, author = {Franco-Barranco, Daniel and Gonz{\'a}lez-Marfil, Aitor and Cardona, Albert and Mu{\~n}oz-Barrutia, Arrate and Arganda-Carreras, Ignacio}, booktitle = {2025 IEEE 22nd International Symposium on Biomedical Imaging (ISBI)}, pages = {1--5}, year = {2025}, doi = {10.1109/ISBI60581.2025.10981218}, }
Nat Methods
BiaPy: Accessible deep learning on bioimages

Daniel Franco-Barranco, Jesús A Andrés-San Román, Ivan Hidalgo-Cenalmor, and 9 more authors

Nature Methods, 2025

Abs DOI Bib HTML PDF Code

BiaPy is an open-source library and application that streamlines the use of common deep learning approaches for bioimage analysis. Designed to simplify technical complexities, it offers an intuitive interface, zero-code notebooks, and Docker integration, catering to both users and developers. While focused on deep learning workflows for 2D and 3D image data, it enhances performance with multi-GPU capabilities, memory optimization, and scalability for large datasets. Although BiaPy does not encompass all aspects of bioimage analysis, such as visualization and manual annotation tools, it empowers researchers by providing a ready-to-use environment with customizable templates that facilitate sophisticated bioimage analysis workflows.
@article{franco2025biapy, title = {BiaPy: Accessible deep learning on bioimages}, author = {Franco-Barranco, Daniel and Andr{\'e}s-San Rom{\'a}n, Jes{\'u}s A and Hidalgo-Cenalmor, Ivan and Backov{\'a}, Lenka and Gonz{\'a}lez-Marfil, Aitor and Caporal, Cl{\'e}ment and Chessel, Anatole and G{\'o}mez-G{\'a}lvez, Pedro and Escudero, Luis M and Wei, Donglai and Mu{\~n}oz-Barrutia, Arrate and Arganda-Carreras, Ignacio}, journal = {Nature Methods}, year = {2025}, pages = {1--3}, publisher = {Nature Publishing Group US New York}, doi = {10.1038/s41592-025-02699-y}, }

2024

Nat Mach Intell
A deep learning method that identifies cellular heterogeneity using nanoscale nuclear features

Davide Carnevali, Limei Zhong, Esther González-Almela, and 9 more authors

Nature Machine Intelligence, Sep 2024

Abs DOI Bib

Cellular phenotypic heterogeneity is an important hallmark of many biological processes and understanding its origins remains a substantial challenge. This heterogeneity often reflects variations in the chromatin structure, influenced by factors such as viral infections and cancer, which dramatically reshape the cellular landscape. To address the challenge of identifying distinct cell states, we developed artificial intelligence of the nucleus (AINU), a deep learning method that can identify specific nuclear signatures at the nanoscale resolution. AINU can distinguish different cell states based on the spatial arrangement of core histone H3, RNA polymerase II or DNA from super-resolution microscopy images. With only a small number of images as the training data, AINU correctly identifies human somatic cells, human-induced pluripotent stem cells, very early stage infected cells transduced with DNA herpes simplex virus type 1 and even cancer cells after appropriate retraining. Finally, using AI interpretability methods, we find that the RNA polymerase II localizations in the nucleoli aid in distinguishing human-induced pluripotent stem cells from their somatic cells. Overall, AINU coupled with super-resolution microscopy of nuclear structures provides a robust tool for the precise detection of cellular heterogeneity, with considerable potential for advancing diagnostics and therapies in regenerative medicine, virology and cancer biology.
@article{carnevali2024ainu, author = {Carnevali, Davide and Zhong, Limei and Gonz{\'a}lez-Almela, Esther and Viana, Carlotta and Rotkevich, Mikhail and Wang, Aiping and Franco-Barranco, Daniel and Gonzalez-Marfil, Aitor and Neguembor, Maria Victoria and Castells-Garcia, Alvaro and Arganda-Carreras, Ignacio and Cosma, Maria Pia}, title = {A deep learning method that identifies cellular heterogeneity using nanoscale nuclear features}, journal = {Nature Machine Intelligence}, year = {2024}, month = sep, volume = {6}, number = {9}, pages = {1021--1033}, issn = {2522-5839}, doi = {10.1038/s42256-024-00883-x}, url = {https://doi.org/10.1038/s42256-024-00883-x}, }
ISBI
Self-supervised Vision Transformers for image-to-image labeling: a BiaPy solution to the LightMyCells Challenge

Daniel Franco-Barranco, Aitor González-Marfil, and Ignacio Arganda-Carreras

In 2024 IEEE International Symposium on Biomedical Imaging (ISBI), 2024

Abs DOI Bib PDF Code

Fluorescence microscopy plays a crucial role in cellular analysis but is often hindered by phototoxicity and limited spectral channels. Label-free transmitted light microscopy presents an attractive alternative, yet recovering fluorescence images from such inputs remains difficult. In this work, we address the Cell Painting problem within the LightMyCells challenge at the International Symposium on Biomedical Imaging (ISBI) 2024, aiming to predict optimally focused fluorescence images from label-free transmitted light inputs. Leveraging advancements in self-supervised Vision Transformers, our method overcomes the constraints of scarce annotated biomedical data and fluorescence microscopy’s drawbacks. Four specialized models, each targeting a different organelle, are pretrained in a self-supervised manner to enhance model generalization. Our method, integrated within the open-source BiaPy library, contributes to the advancement of image-to-image deep-learning techniques in cellular analysis, offering a promising solution for robust and accurate fluorescence image prediction from label-free transmitted light inputs.
@inproceedings{franco2024lightmycells, author = {Franco-Barranco, Daniel and Gonz{\'a}lez-Marfil, Aitor and Arganda-Carreras, Ignacio}, booktitle = {2024 IEEE International Symposium on Biomedical Imaging (ISBI)}, title = {Self-supervised Vision Transformers for image-to-image labeling: a BiaPy solution to the LightMyCells Challenge}, year = {2024}, pages = {1--5}, doi = {10.1109/ISBI56570.2024.10635818}, }
A&A
Characterizing structure formation through instance segmentation

Daniel López-Cano, Jens Stücker, Marcos Pellejero Ibañez, and 2 more authors

Astronomy & Astrophysics, 2024

Abs arXiv Bib HTML PDF Code

Dark matter haloes form from small perturbations to the almost homogeneous density field of the early universe. Although it is known how large these initial perturbations must be to form haloes, it is rather poorly understood how to predict which particles will end up belonging to which halo. However, it is this process that determines the Lagrangian shape of protohaloes and is therefore essential to understand their mass, spin and formation history. Here, we present a machine-learning framework to learn how the protohalo regions of different haloes emerge from the initial density field. This involves one neural network to distinguish semantically which particles become part of any halo and a second neural network that groups these particles by halo membership into different instances. This instance segmentation is done through the Weinberger method, in which the network maps particles into a pseudo-space representation where different instances can be distinguished easily through a simple clustering algorithm. Our model reliably predicts the masses and Lagrangian shapes of haloes object-by-object, as well as summary statistics like the halo-mass function. We find that our model extracts information close to optimal by comparing it to the degree of agreement between two N-body simulations with slight differences in their initial conditions. We publish our model open-source and suggest that it can be used to inform analytical methods of structure formation by studying the effect of systematic manipulations of the initial conditions.
@article{lopez2024cosmos, title = {Characterizing structure formation through instance segmentation}, author = {L{\'o}pez-Cano, Daniel and St{\"u}cker, Jens and Iba{\~n}ez, Marcos Pellejero and Angulo, Ra{\'u}l E and Franco-Barranco, Daniel}, journal = {Astronomy \& Astrophysics}, volume = {685}, pages = {A37}, year = {2024}, publisher = {EDP Sciences}, }

2023

ISBI
Modeling Wound Healing Using Vector Quantized Variational Autoencoders and Transformers

Lenka Backová, Guillermo Bengoetxea, Svana Rogalla, and 3 more authors

In 2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI), 2023

Abs Bib HTML PDF Code

Wound healing is a fundamental mechanism for living animals. Understanding the process is crucial for numerous medical applications ranging from scarless healing to faster tissue regeneration and safer post-surgery recovery. In this work, we collect a dataset of time-lapse sequences of Drosophila embryos recovering from a laser-incised wound. We model the wound healing process as a video prediction task for which we utilize a two-stage approach with a vector quantized variational autoencoder and an autoregressive transformer. We show our trained model is able to generate realistic videos conditioned on the initial frames of the healing. We evaluate the model predictions using distortion measures and perceptual quality metrics based on segmented wound masks. Our results show that the predictions keep pixel-level error low while behaving in a realistic manner, thus suggesting the neural network is able to model the wound-closing process.
@inproceedings{backova2023woundhealing, title = {Modeling Wound Healing Using Vector Quantized Variational Autoencoders and Transformers}, author = {Backov{\'a}, Lenka and Bengoetxea, Guillermo and Rogalla, Svana and Franco-Barranco, Daniel and Solon, J{\'e}r{\o}me and Arganda-Carreras, Ignacio}, booktitle = {2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI)}, pages = {1--5}, year = {2023}, organization = {IEEE}, }
ISBI
BiaPy: a ready-to-use library for Bioimage Analysis Pipelines

Daniel Franco-Barranco, Jesús A Andrés-San Román, Pedro Gómez-Gálvez, and 3 more authors

In 2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI), 2023

Abs Bib HTML PDF Code

In recent years, technological advances in microscopy have made available large amounts of data to biomedical researchers in the form of images. By learning from such large datasets, deep learning-based methods have successfully addressed previously inaccessible bioimage analysis tasks. However, most available solutions target a particular subset of problems, forcing users to be familiarized with different applications to complete their data analysis. On top of that, other issues, such as reproducibility, lack of documentation, or access to the code, arise. For these reasons, we introduce BiaPy, an open-source ready-to-use all-in-one library that provides deep-learning workflows for a large variety of bioimage analysis tasks, including 2D and 3D semantic and instance segmentation, object detection, super-resolution, denoising, self-supervised learning, and classification.
@inproceedings{franco2023biapy, title = {BiaPy: a ready-to-use library for Bioimage Analysis Pipelines}, author = {Franco-Barranco, Daniel and Andr{\'e}s-San Rom{\'a}n, Jes{\'u}s A and G{\'o}mez-G{\'a}lvez, Pedro and Escudero, Luis M and Mu{\~n}oz-Barrutia, Arrate and Arganda-Carreras, Ignacio}, booktitle = {2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI)}, pages = {1--5}, year = {2023}, organization = {IEEE}, }
IEEE TMI
Current Progress and Challenges in Large-scale 3D Mitochondria Instance Segmentation

Daniel Franco-Barranco, Zudi Lin, Won-Dong Jang, and 24 more authors

IEEE Transactions on Medical Imaging, 2023

Abs Bib HTML PDF Code

In this paper, we present the results of the MitoEM challenge on mitochondria 3D instance segmentation from electron microscopy images, organized in conjunction with the IEEE-ISBI 2021 conference. Our benchmark dataset consists of two large-scale 3D volumes, one from human and one from rat cortex tissue, which are 1,986 times larger than previously used datasets. At the time of paper submission, 257 participants had registered for the challenge, 14 teams had submitted their results, and six teams participated in the challenge workshop. Here, we present eight top-performing approaches from the challenge participants, along with our own baseline strategies. Posterior to the challenge, annotation errors in the ground truth were corrected without altering the final ranking. Additionally, we present a retrospective evaluation of the scoring system which revealed that: 1) challenge metric was permissive with the false positive predictions; and 2) size-based grouping of instances did not correctly categorize mitochondria of interest. Thus, we propose a new scoring system that better reflects the correctness of the segmentation results. Although several of the top methods are compared favorably to our own baselines, substantial errors remain unsolved for mitochondria with challenging morphologies.
@article{franco2023current, title = {Current Progress and Challenges in Large-scale 3D Mitochondria Instance Segmentation}, author = {Franco-Barranco, Daniel and Lin, Zudi and Jang, Won-Dong and Wang, Xueying and Shen, Qijia and Yin, Wenjie and Fan, Yutian and Li, Mingxing and Chen, Chang and Xiong, Zhiwei and Xin, Rui and Liu, Hao and Chen, Huai and Li, Zhili and Zhao, Jie and Chen, Xuejin and Pape, Constantin and Conrad, Ryan and De Folter, Jozefus and Nightingale, Luke and Jones, Martin and Liu, Yanling and Ziaei, Dorsa and Huschauer, Stephan and Arganda-Carreras, Ignacio and Pfister, Hanspeter and Wei, Donglai}, journal = {IEEE Transactions on Medical Imaging}, year = {2023}, }
Cell Rep Methods
CartoCell, a high-content pipeline for 3D image analysis, unveils cell morphology patterns in epithelia

Jesus A Andres-San Roman, Carmen Gordillo-Vazquez, Daniel Franco-Barranco, and 10 more authors

Cell Reports Methods, 2023

Abs Bib HTML PDF Code

Decades of research have not yet fully explained the mechanisms of epithelial self-organization and 3D packing. Single-cell analysis of large 3D epithelial libraries is crucial for understanding the assembly and function of whole tissues. Combining 3D epithelial imaging with advanced deep-learning segmentation methods is essential for enabling this high-content analysis. We introduce CartoCell, a deep-learning-based pipeline that uses small datasets to generate accurate labels for hundreds of whole 3D epithelial cysts. Our method detects the realistic morphology of epithelial cells and their contacts in the 3D structure of the tissue. CartoCell enables the quantification of geometric and packing features at the cellular level. Our single-cell cartography approach then maps the distribution of these features on 2D plots and 3D surface maps, revealing cell morphology patterns in epithelial cysts. Additionally, we show that CartoCell can be adapted to other types of epithelial tissues.
@article{andres2023cartocell, title = {CartoCell, a high-content pipeline for 3D image analysis, unveils cell morphology patterns in epithelia}, author = {Andres-San Roman, Jesus A and Gordillo-Vazquez, Carmen and Franco-Barranco, Daniel and Morato, Laura and Fernandez-Espartero, Cecilia H and Baonza, Gabriel and Tagua, Antonio and Vicente-Munuera, Pablo and Palacios, Ana M and Gavil{\'a}n, Mar{\'\i}a P and G{\'o}mez-G{\'a}lvez, Pedro and Arganda-Carreras, Ignacio and Escudero, Luis M}, journal = {Cell Reports Methods}, volume = {3}, number = {10}, year = {2023}, publisher = {Elsevier}, }

2022

CMPB
Deep learning based domain adaptation for mitochondria segmentation on EM volumes

Daniel Franco-Barranco, Julio Pastor-Tronch, Aitor Gonzalez-Marfil, and 2 more authors

Computer Methods and Programs in Biomedicine, 2022

Abs DOI Bib PDF Supp Code

Background and Objective: Accurate segmentation of electron microscopy (EM) volumes of the brain is essential to characterize neuronal structures at a cell or organelle level. While supervised deep learning methods have led to major breakthroughs in that direction during the past years, they usually require large amounts of annotated data to be trained, and perform poorly on other data acquired under similar experimental and imaging conditions. This is a problem known as domain adaptation, since models that learned from a sample distribution (or source domain) struggle to maintain their performance on samples extracted from a different distribution or target domain. In this work, we address the complex case of deep learning based domain adaptation for mitochondria segmentation across EM datasets from different tissues and species. Methods: We present three unsupervised domain adaptation strategies to improve mitochondria segmentation in the target domain based on (1) state-of-the-art style transfer between images of both domains; (2) self-supervised learning to pre-train a model using unlabeled source and target images, and then fine-tune it only with the source labels; and (3) multi-task neural network architectures trained end-to-end with both labeled and unlabeled images. Additionally, to ensure good generalization in our models, we propose a new training stopping criterion based on morphological priors obtained exclusively in the source domain. Results: We carried out all possible cross-dataset experiments using three publicly available EM datasets. We evaluated our proposed strategies and those of others based on the mitochondria semantic labels predicted on the target datasets. Conclusions: The methods introduced here outperform the baseline methods and compare favorably to the state of the art.
@article{francobarranco2022domain, title = {Deep learning based domain adaptation for mitochondria segmentation on EM volumes}, journal = {Computer Methods and Programs in Biomedicine}, pages = {106949}, year = {2022}, issn = {0169-2607}, doi = {10.1016/j.cmpb.2022.106949}, author = {Franco-Barranco, Daniel and Pastor-Tronch, Julio and Gonzalez-Marfil, Aitor and Mu{\~n}oz-Barrutia, Arrate and Arganda-Carreras, Ignacio}, }
Cell Systems
A quantitative biophysical principle to explain the 3D cellular connectivity in curved epithelia

Pedro Gómez-Gálvez, Pablo Vicente-Munuera, Samira Anbari, and 14 more authors

Cell Systems, 2022

Preprint DOI shown; see html link for the final Cell Systems version.

Abs DOI Bib HTML PDF

Epithelial cell organization and the mechanical stability of tissues are closely related. In this context, it has been recently shown that packing optimization in bended/folded epithelia is achieved by a surface tension energy minimization mechanism that leads to a novel cellular shape: the scutoid. However, further cellular and tissue level implications of this new developmental paradigm remain unknown. Here we focus on the relationship between this complex cellular shape and the connectivity between cells. We address this problem using a combination of computational, experimental, and biophysical approaches in tubular epithelia. In particular, we examine how energy drivers affect the three-dimensional packing of these tissues. We challenge our biophysical model by reducing the cell adhesion in epithelial cells. As a result, we observed an increment on the cell apico-basal intercalation propensity that correlated with a decrease of the energy barrier necessary to connect with new cells. We conclude that tubular epithelia satisfy a quantitative biophysical principle, that links tissue geometry and energetics with the average cellular connectivity.
@article{gomezgalvez2022epithelia, author = {G{\'o}mez-G{\'a}lvez, Pedro and Vicente-Munuera, Pablo and Anbari, Samira and Tagua, Antonio and Gordillo-V{\'a}zquez, Carmen and Andr{\'e}s-San Rom{\'a}n, Jes{\'u}s A. and Franco-Barranco, Daniel and Palacios, Ana M. and Velasco, Antonio and Capit{\'a}n-Agudo, Carlos and Grima, Clara and Annese, Valentina and Arganda-Carreras, Ignacio and Robles, Rafael and M{\'a}rquez, Alberto and Buceta, Javier and Escudero, Luis M.}, title = {A quantitative biophysical principle to explain the 3D cellular connectivity in curved epithelia}, journal = {Cell Systems}, year = {2022}, doi = {10.1101/2020.02.19.955567}, note = {Preprint DOI shown; see html link for the final Cell Systems version.}, }

Book

Building a Bioimage Analysis Workflow Using Deep Learning

Estibaliz Gómez-de-Mariscal, Daniel Franco-Barranco, Arrate Muñoz-Barrutia, and 1 more author

In Bioimage Data Analysis Workflows – Advanced Components and Methods, 2022

Bib HTML Code

@incollection{gomezdemariscal2021building,
  title = {Building a Bioimage Analysis Workflow Using Deep Learning},
  author = {G{\'o}mez-de-Mariscal, Estibaliz and Franco-Barranco, Daniel and Mu{\~n}oz-Barrutia, Arrate and Arganda-Carreras, Ignacio},
  booktitle = {Bioimage Data Analysis Workflows -- Advanced Components and Methods},
  year = {2022},
  publisher = {Springer},
}

2021

Neuroinformatics
Stable Deep Neural Network Architectures for Mitochondria Segmentation on Electron Microscopy Volumes

Daniel Franco-Barranco, Arrate Muñoz-Barrutia, and Ignacio Arganda-Carreras

Neuroinformatics, Dec 2021

Abs DOI arXiv Bib PDF Code

Electron microscopy (EM) allows the identification of intracellular organelles such as mitochondria, providing insights for clinical and scientific studies. In recent years, a number of novel deep learning architectures have been published reporting superior performance, or even human-level accuracy, compared to previous approaches on public mitochondria segmentation datasets. Unfortunately, many of these publications do not make neither the code nor the full training details public to support the results obtained, leading to reproducibility issues and dubious model comparisons. For that reason, and following a recent code of best practices for reporting experimental results, we present an extensive study of the state-of-the-art deep learning architectures for the segmentation of mitochondria on EM volumes, and evaluate the impact in performance of different variations of 2D and 3D U-Net-like models for this task. To better understand the contribution of each component, a common set of pre- and post-processing operations has been implemented and tested with each approach. Moreover, an exhaustive sweep of hyperparameters values for all architectures have been performed and each configuration has been run multiple times to report the mean and standard deviation values of the evaluation metrics. Using this methodology, we found very stable architectures and hyperparameter configurations that consistently obtain state-of-the-art results in the well-known EPFL Hippocampus mitochondria segmentation dataset. Furthermore, we have benchmarked our proposed models on two other available datasets, Lucchi++ and Kasthuri++, where they outperform all previous works.
@article{francobarranco2021stable, author = {Franco-Barranco, Daniel and Mu{\~{n}}oz-Barrutia, Arrate and Arganda-Carreras, Ignacio}, title = {Stable Deep Neural Network Architectures for Mitochondria Segmentation on Electron Microscopy Volumes}, journal = {Neuroinformatics}, year = {2021}, month = dec, issn = {1559-0089}, doi = {10.1007/s12021-021-09556-1}, url = {https://doi.org/10.1007/s12021-021-09556-1}, }

2020

MICCAI
MitoEM dataset: Large-scale 3D mitochondria instance segmentation from EM images

Donglai Wei, Zudi Lin, Daniel Franco-Barranco, and 10 more authors

In International Conference on Medical Image Computing and Computer-Assisted Intervention, 2020

Abs Bib PDF Code

Electron microscopy (EM) allows the identification of intracellular organelles such as mitochondria, providing insights for clinical and scientific studies. However, public mitochondria segmentation datasets only contain hundreds of instances with simple shapes. It is unclear if existing methods achieving human-level accuracy on these small datasets are robust in practice. To this end, we introduce the MitoEM dataset, a 3D mitochondria instance segmentation dataset with two (30\textmum)^3 volumes from human and rat cortices respectively, 3,600\texttimes larger than previous benchmarks. With around 40K instances, we find a great diversity of mitochondria in terms of shape and density. For evaluation, we tailor the implementation of the average precision (AP) metric for 3D data with a 45\texttimes speedup. On MitoEM, we find existing instance segmentation methods often fail to correctly segment mitochondria with complex shapes or close contacts with other instances. Thus, our MitoEM dataset poses new challenges to the field.
@inproceedings{wei2020mitoem, title = {MitoEM dataset: Large-scale 3D mitochondria instance segmentation from EM images}, author = {Wei, Donglai and Lin, Zudi and Franco-Barranco, Daniel and Wendt, Nils and Liu, Xingyu and Yin, Wenjie and Huang, Xin and Gupta, Aarush and Jang, Won-Dong and Wang, Xueying and Arganda-Carreras, Ignacio and Lichtman, Jeff W. and Pfister, Hanspeter}, booktitle = {International Conference on Medical Image Computing and Computer-Assisted Intervention}, pages = {66--76}, year = {2020}, organization = {Springer}, }