Daniel Franco-Barranco

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.

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.

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.

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.

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.

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.

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.

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.

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.