Biology subjects

Delgado-Rodriguez, P.

Publications and source records attributed to Delgado-Rodriguez, P..

3 recordsLinked to original sources

bioRxiv2025

In-depth 3D Exploration of Autosomal Dominant Polycystic Kidney Disease Through Light Sheet Fluorescence Microscopy

Autosomal Dominant Polycystic Kidney Disease (ADPKD) is the most prevalent genetic kidney disorder. Animal preclinical studies are one of the main tools to study this disease, often through either 2D histology imaging for high-resolution analysis or CT or MRI for full kidney segmentation. As an alternative to these modalities, we propose the use of Light Sheet Fluorescence Microscopy (LSFM) for high-resolution 3D imaging of healthy and ADPKD-induced mouse kidneys, enabling a detailed volumetric morphological analysis of the diseases effects. In a mouse ADPKD model, ex vivo imaging of the kidneys was performed through LSFM, after which a combination of machine learning and other processing techniques allowed us to perform an in-depth image analysis. This includes the segmentation of key structures, such as the full kidney volume and, within it, its internal cavities, cortex, glomeruli, and cysts, complemented by texture analysis of tubular structures in the cortical area. Pathological kidneys exhibited significant volume enlargement and increased internal cavities due to cystogenesis. While glomerular count remained stable, their spatial distribution was altered, showing increased interglomerular distances and show-casing the deformations produced by the disease. The texture analysis of tubules from the cortex region identified Local Binary Pattern (LBP) uniformity and porosity as key biomarkers of tissue deformation, which could be used as markers to further evaluate the development of the disease. These findings underscore the potential of LSFM imaging as a powerful tool for detailed ADPKD characterization and treatment assessment.

bioengineering↗

bioRxiv2024

SAM-based Automatic Workflow for Histology Cyst Segmentation in Autosomal Dominant Polycystic Kidney Disease

Autosomal Dominant Polycystic Kidney Disease (ADPKD) is a genetic disorder characterized by the development of numerous cysts in the kidneys, ultimately leading to significant structural alterations and renal failure. Detailed investigations of this disease frequently utilize histological analyses of kidney sections across various stages of ADPKD progression. In this paper, we introduce an automated workflow leveraging the Segment Anything Model (SAM) neural network, complemented by a series of post-processing steps, to autonomously segment cysts in histological images. This approach eliminates the need for manual annotations or preliminary training phases and enables precise quantification of cystic changes over entire kidney sections. Application of this method to sequential histology images across the development timeline of ADPKD in mice demonstrated a notable increase in the proportion of diseased tissue from 8 to 12 weeks and from 12 to 16 weeks, with the cysts appearing progressively lighter. Our workflow not only surpasses the performance of the existing Cystanalyser tool but also offers enhanced flexibility and accuracy in full-image segmentation. The developed workflow is made publicly accessible to facilitate its adoption as an efficient tool for rapid and reliable cyst segmentation in histological studies.

bioinformatics↗

bioRxiv2023

Automatic Classification of Normal and Abnormal Cell Division Using Deep Learning Analysis of Mitosis Videos

In recent years, there has been a surge in the development of methods for cell segmentation and tracking, with initiatives such as the Cell Tracking Challenge driving progress in the field. Most studies focus on regular cell population videos in which cells are segmented, cell tracks followed, and parental relationships annotated. However, DNA damage induced by genotoxic drugs or ionizing radiation provide additional abnormal cellular events of interest since they lead to aberrant behaviors such as abnormal cell divisions (i.e., resulting in a number of daughter cells different from two) and cell death. The dynamic development of those abnormal events can be followed using time lapse microscopy to be further analyzed. With this in mind, we developed an automatic mitosis classifier that categorizes small mitosis image sequences centered around a single cell as "Normal" or "Abnormal." These mitosis sequences were extracted from videos of cell populations exposed to varying levels of radiation that affect the cell cycles development. Such an approach can aid in detecting, tracking, and characterizing the behavior of the entire population. In this study, we explored several deep-learning architectures for working with 12-frame mitosis sequences. We found that a network with a ResNet50 backbone, modified to operate independently on each video frame and then combined using a Long Short-Term Memory (LSTM) layer, produced the best results in the classification (mean F1-score: 0.93 {+/-} 0.06). In future work, we plan to integrate the mitosis classifier in a cell segmentation and tracking pipeline to build phylogenetic trees of the entire cell population after genomic stress. Author SummaryIn recent years, there has been a growing interest in developing methods to analyze videos of cell populations, which show how cells move and divide over time. Typically, researchers focus on developing methods to automatically identify and track individual cells and their divisions. However, exposure to anticancer drugs or radiation can cause uncommon behaviors, such as abnormal cell divisions, which are of interest to experts studying the effects of these agents on cell behavior. To address this issue, we developed an automated tool that can determine whether a specific cell division seen in a video is normal or abnormal. We used video microscopy to capture small sequences of cell division, and then trained a deep-learning model to classify these sequences as either normal or abnormal. We found that our model achieved a high level of accuracy in this task. Our tool has the potential to aid experts in identifying abnormal cellular events, providing insights into the effects of genotoxic agents on cell behavior. In future work, we plan to integrate our tool into more complex methods for analyzing cell population videos, which may help us better understand the impact of toxic agents on the behavior of the entire cell population.

bioinformatics↗