AlveolEye: Rapid and precise lung morphometry guided by computer vision

Hirsh, J.; Hirsh, S.; Shirazi, S. P.; Douglas, I.; Garg, S.; Son, Y.; Pierre-Louis, A.; Bunn, C.; Jetter, C. S.; James, T. J.; Sharkey, A. L.; Benjamin, J. T.; Kropski, J. A.; Sucre, J. M. S.; Negretti, N. M.

doi:10.1101/2025.09.30.679300

bioRxiv · 10.1101/2025.09.30.679300

AlveolEye: Rapid and precise lung morphometry guided by computer vision

Hirsh, J.·Hirsh, S.·Shirazi, S. P.·Douglas, I.·Garg, S.·Son, Y.·Pierre-Louis, A.·Bunn, C.·Jetter, C. S.·James, T. J.·Sharkey, A. L.·Benjamin, J. T.·Kropski, J. A.·Sucre, J. M. S.·Negretti, N. M.

View original source ↗Export metadata ↓

Abstract

Rigorous and reproducible evaluation of lung tissue under different conditions is necessary to interpret development, injury, and pharmacologic interventions. Common histological measurements in the distal lung include mean linear intercept (MLI) as a metric of alveolarization and airspace volume density (ASVD) as a metric of airspaces relative to tissue. Historically, these have been performed manually in a time-intensive process, with reproducible trends, but a high degree of variability between individuals. To improve the reproducibility and throughput of lung morphometry, we developed AlveolEye, an open source, semi-automated, computer vision-assisted tool that rapidly and reproducibly calculates MLI and ASVD from images of standard hematoxylin and eosin (H&E) stained tissue sections. AlveolEye-assisted MLI calculation closely aligns with manually-derived measurements for corresponding images, with preservation of trends in measurements between non-injured controls and neonatal mice subjected to two different injury models. Analyzing human tissue of varying ages suggests that the approach developed in AlveolEye is generalizable across species. Notably, AlveolEye markedly reduced the average variation across individual analyzers, with the greatest improvement in precision among individuals with the least experience in performing lung morphometry. The design of AlveolEye is intentionally semi-automated, preserving the investigators ability to assess and adjust parameters based on sample characteristics. AlveolEye facilitates efficient lung morphological measurements on larger sample sizes, allowing for greater statistical power for preclinical studies, and improves precision across individual observers, allowing for improved rigor in experimental design and execution.

Source connections

Published article ↗

Explore related subjects

developmental biology Preprint

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Hirsh, J., Hirsh, S., Shirazi, S. P., Douglas, I., Garg, S., Son, Y., Pierre-Louis, A., Bunn, C., Jetter, C. S., James, T. J., Sharkey, A. L., Benjamin, J. T., Kropski, J. A., Sucre, J. M. S., Negretti, N. M.. 2025-10-02. AlveolEye: Rapid and precise lung morphometry guided by computer vision. https://doi.org/10.1101/2025.09.30.679300

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

Hyaluronic acid dynamics are regulated by neural crest cells and NEDD4 in the developing cardiac outflow tract valves

The extracellular matrix (ECM) is essential for cardiac valve development, with hyaluronic acid (HA) playing a critical role in cardiac cushion formation. While HA is required for atrioventricular valve development, its role in semilunar valve formation remains poorly understood. The ubiquitin ligase NEDD4 is essential for cardiac outflow tract development, particularly within cardiac neural crest cells, however its role in semilunar valve development has been unexplored. Here, we demonstrate that neural crest-specific deletion of NEDD4 in mice disrupts outflow tract valve development, resulting in abnormal leaflet number and morphology. Analysis of valve ECM revealed a specific reduction in HA within the neural crest-derived valve mesenchyme, while other ECM components such as fibronectin and versican remained unaffected. In contrast, deletion of NEDD4 in endocardium-derived valve mesenchyme did not alter HA abundance. Using NEDD4-deficient neural crest cells, we further demonstrate a cell-autonomous requirement for NEDD4 in regulating HA abundance, with in vitro evidence supporting impaired HA synthesis rather than increased turnover of HA in the ECM. Our findings identify a previously unrecognised role for NEDD4 in regulating HA synthesis in cardiac neural crest cells, revealing a cell-type-specific mechanism governing semilunar valve development and highlighting distinct cellular requirements for ECM regulation across cardiac valve populations.

developmental biology↗

bioRxiv2026

Cxcr2 is required for ductal outgrowth and shapes the stromal microenvironment in postnatal mammary gland development

Mammary gland development requires coordinated signaling between epithelial and stromal compartments. While CXC chemokine receptor 2 (CXCR2) is known to regulate inflammatory and tissue remodeling processes in disease, including breast cancer, its role in mammary gland morphogenesis is not well understood. Here, we examined postnatal mammary gland development in germline Cxcr2 knockout mice. Loss of Cxcr2 resulted in delayed ductal outgrowth, reduced epithelial area, and altered stromal architecture, including decreased adipocyte size and increased collagen deposition, despite preserved branching organization. These defects occurred in the context of severe systemic immune dysregulation, marked by expansion of granulocytic populations and loss of adaptive immune cells. Together, our findings indicate that CXCR2 indirectly regulates mammary gland development through coordination between the epithelium and stroma, rather than acting through epithelial cell-intrinsic mechanisms. Because the same myeloid expansion, immune suppression, and collagen-rich stromal remodeling that CXCR2 controls during normal development are hallmarks of the pro-tumorigenic breast microenvironment, these findings establish CXCR2 as a regulator of the immune and stromal landscape of the mammary gland and provide a developmental framework for understanding its contributions to the breast tumor microenvironment.

developmental biology↗

bioRxiv2026

Endogenous Mechanical Reporters Encode the Global Mechanical Landscape of Morphogenesis

Developmental systems usually comprise multiple genetically programmed morphogenetic events within a mechanically continuous environment, yet how individual processes interact and contribute to the overall mechanical landscape remains unknown. Here, we introduce the concept of mechanical reporter: endogenous tissues function as in situ probes of the embedded mechanical landscape. We exemplified cephalic furrow (CF) migration as a mechanical reporter in gastrulating Drosophila. Using optogenetics and genetics, we revealed that CF displacement reflects the combined effect of both force and tissue mechanical properties. By incorporating condition-specific mechanical landscape into a continuum minimal-response model, we developed a framework to quantitatively predict the reporter dynamics. Using different genetic mutants, we successfully decoded changes in the mechanical landscape from CF dynamics in gastrulating Drosophila. Overall, we established a predictive framework that uses CF dynamics to infer changes in the global mechanical landscape. Beyond this, the concept of mechanical reporter provides a general strategy for systematically understanding and decoding complex mechanical environments in living systems.

developmental biology↗