VISTA-Z: Vascular Imaging and Segmentation for Topology Analysis in Zebrafish

VISTA-Z: Vascular Imaging and Segmentation for Topology Analysis in Zebrafish | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article VISTA-Z: Vascular Imaging and Segmentation for Topology Analysis in Zebrafish Ignacio Rodriguez-Pastrana, Joanna Richens, Robert N. Wilkinson This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8402098/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 01 Apr, 2026 Read the published version in Scientific Reports → Version 1 posted 10 You are reading this latest preprint version Abstract High-resolution vascular imaging in zebrafish embryos offers unparalleled insight into angiogenesis, yet quantitative analysis remains limited by manual workflows and inconsistent segmentation. To overcome this, we developed VISTA-Z, an automated Python-based pipeline designed to standardise vascular quantification across diverse fluorescent datasets. The workflow combines adaptive contrast enhancement, vessel segmentation using Meijering filtering, artefact removal through segment labelling, and skeleton-based topology analysis with refined branchpoint detection. Metrics are normalised to physical units and imaging depth, enabling reproducible comparisons across experiments. We validated VISTA-Z using multiple endothelial transgenic lines and developmental stages. The pipeline detected subtle and severe phenotypes, including brain vessel loss in foxc1a mutants, brain and trunk vessel loss in kdrl mutants and widespread hyper-angiogenesis in plxnd1 crispants. VISTA-Z is an open-source, scalable platform for reproducible high-throughput quantification of zebrafish vascular architecture, providing a standardised framework for developmental research and preclinical screening. Biological sciences/Biological techniques Biological sciences/Biotechnology Biological sciences/Computational biology and bioinformatics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Cardiovascular disease remains the leading cause of mortality worldwide, with vascular dysfunction playing a central role in its pathogenesis [1]. This burden underscores the importance of vascular imaging in both cardiovascular medicine and preclinical research. Vascular imaging provides structural information, including vessel morphology and lumen diameter, that influences blood flow dynamics (haemodynamics) [2]. Applying vascular imaging in preclinical models enables mechanistic studies of angiogenesis, vascular remodelling, and haemodynamic regulation under physiological and pathological conditions [3]. Such insights are essential to advance our understanding of vascular disorders. Zebrafish ( Danio rerio ) is a powerful vertebrate model for studying vascular development and cardiovascular disease [4]. Around 70% of human genes have zebrafish orthologs, including 82% of known disease genes, making it highly relevant for modelling human pathophysiology [5]. The optical transparency of zebrafish embryos and the ease of generating vascular-specific fluorescent transgenic lines enable real-time in vivo imaging of the entire vascular network. Combined with external fertilisation, rapid ex-utero development, and the capability to survive without circulation for several days, the zebrafish allows the study of gene function leading to severe vascular phenotypes that are lethal in mammals [6]. These advantages position zebrafish as a key model to investigate angiogenesis, vessel architecture, and vascular malformations [7]. Despite advances in imaging, zebrafish vascular imaging often falls short of its full potential because manual analysis remains slow, subjective and error prone. Automated pipelines overcome these barriers by delivering rapid, reproducible, and unbiased quantification of vascular metrics [8]. Historically, most vascular analysis tools were designed for datasets with high signal-to-noise ratios and uniform contrast, such as the retinal vasculature [9]. Several algorithms developed for zebrafish vascular analysis address challenges posed by heterogeneous fluorescence and low signal-to-noise ratios [8,10,11]. While initial approaches to zebrafish vascular quantification relied on microangiography and manual measurements of selected vessels, this required extensive user input and offered limited scalability [12]. To overcome these constraints, more recent methodologies have introduced automated pipelines for processing fluorescent imaging data. Several of these new algorithms are available through open-source platforms or incorporate graphical user interfaces to improve accessibility and streamline analysis [8,10,11]. Although these developments represent substantial progress, many remain confined to specific anatomical regions or require considerable manual correction. These limitations underscore the need for fully automated, scalable solutions capable of accurately quantifying vascular structures across diverse imaging datasets. To overcome the limitations of existing approaches, we developed VISTA-Z, an automated pipeline for zebrafish vasculature quantification. This workflow integrates unsupervised contrast adjustment, thresholding for vessel background separation, and filtering techniques that highlight tubular structures to achieve robust segmentation. In addition, VISTA-Z introduces novel functionalities, including segment labelling for manual artefact removal, skeletonisation with branchpoint refinement, and normalisation of metrics to physical units and z-stack depth. These refinements improve biological interpretability by correcting for image and experimental variability. Using VISTA-Z, we quantified the vascular expansion of wild type embryos from 3 to 5 days post-fertilisation (dpf), encompassing both brain and trunk regions. Furthermore, we demonstrate its versatility in detecting vascular abnormalities, such as vessel loss in foxc1a and kdrl mutants, and excessive angiogenesis in plxnd1 crispants. Collectively, VISTA-Z provides a comprehensive, high-throughput framework for vascular phenotyping in embryonic zebrafish models. Results Development of VISTA-Z, an automated pipeline for vascular quantification VISTA-Z was developed to enable automated, unbiased quantification of zebrafish vascular metrics from fluorescent data. It is optimised to handle low signal-to-noise ratios and heterogeneous expression patterns (Fig. 1a). The pipeline begins with automated image pre-processing, including contrast enhancement using CLAHE and vessel masking via Otsu thresholding (Fig. 1b). CLAHE locally improves vessel visibility in low-signal regions without amplifying background noise [13], while Otsu thresholding provides consistent, unbiased masking across samples [8]. Vessel structures are then segmented using Meijering filtering (Fig. 1c), selected for its superior performance in preserving vessel connectivity and enhancing complex vascular features [14]. To address the detection of vessel artefacts, our pipeline assigns a unique identifier to each vessel segment, applies a size-based exclusion and allows manual curation, enabling users to remove specific segments if required (Fig. 1c). Structural metrics are extracted through a two-step skeletonisation and quantification workflow (Fig. 1d). Expanding core functionality found within the VesselMetrics package [11], vessel structures are first converted into single-pixel centreline representations. These skeletons preserve the topology of the vascular network enabling quantification of structural metrics such as vessel length and diameter (Fig. 1d). We next implemented custom modules for refined branchpoint filtering (Fig. 1d), followed by metric extraction, converted measurements to spatial units (microns) and normalised them by z-stack depth (Fig. 1e). These improvements ensure consistency across imaging conditions and enhance the biological interpretability of results. Finally, outliers are removed using MAD filtering (Fig. 1e), reducing the influence of extreme values [15]. VISTA-Z supports confocal images with optional region-of-interest (ROI) selection to accommodate diverse imaging conditions (Fig. 1b’), and code is available via GitHub (https://github.com/fishyvessels/VISTA-Z). Validation of VISTA-Z performance using double transgenic zebrafish embryos The contrast-to-noise ratio varies between vascular-specific fluorescent transgenes, but remains relatively consistent across developmental stages, anatomical regions, and vessel sizes [8]. The use of distinct transgenes provides complementary vascular information that improves segmentation accuracy [16]. It is therefore necessary to validate image analysis algorithms across different fluorescent transgenes and developmental stages to ensure robustness. To validate our pipeline, we employed Tg(kdrl:EGFP) s843 ; Tg(kdrl:Hsa.HRAS-mCherry) s916 double transgenic zebrafish embryos, which fluorescently label the cytoplasm and membrane of endothelial cells, respectively. We performed validation across two anatomically distinct vascular areas, the brain and the trunk, at 3 and 5 dpf to account for both anatomical and developmental changes. At 3 dpf, confocal imaging of the brain vasculature revealed comparable network architecture between both reporters (Fig. 2a). No significant differences in normalised network length (Fig. 2b), vessel density (Fig. 2c), and number of branchpoints (Fig. 2d) were observed between the two transgenes. Interestingly, vessel diameters were significantly lower when measured using the Tg(kdrl:EGFP) s843 transgene compared to Tg(kdrl:Hsa.HRAS-mCherry) s916 (Fig. 2e). Linear regression showed a moderate correlation between diameter measurements (Fig. 2f), suggesting differences between transgenes contributed to observed discrepancies in vessel diameter. A non-uniform cytoplasmic localisation of EGFP could be observed on confocal images (Fig. 2a, S1a-b), unlike the membrane localisation of mCherry [17]. Therefore, to determine whether fluorophore localisation could influence diameter measurements, we examined the cellular localisation of EGFP in a Tg(kdrl:EGFP) s843 ; Tg(fli1a:nls-mCherry) sh550 double transgenic zebrafish line. Co-localisation analysis revealed EGFP localisation within the endothelial cell cytoplasm and nucleus (Fig. S1b-c), as previously shown but not reported [8,18,19]. Collectively, this suggests reduced vessel diameters result from differences cytoplasmic EGFP localisation compared to membrane mCherry localisation. Given that the membrane is the outermost part of the cell and that measurements extending to the membrane would be expected to be greater than those extending to the limit of cytoplasm, this suggests VISTA-Z is sufficiently robust to detect small changes in diameter due to differences in fluorophore localisation in distinct transgenes. We next employed VISTA-Z to capture the stereotypical trunk vascular architecture at 3 dpf (Fig. 2g). Interestingly, normalised network length (Fig. 2h), vessel density (Fig. 2i) and the number of branchpoints (Fig. 2j) were significantly increased using Tg(kdrl:EGFP) s843 compared to Tg(kdrl:Hsa.HRAS-mCherry) s916 . Tg(kdrl:EGFP) s843 labelled the developing sub-intestinal vein plexus (SIVP) in contrast to Tg(kdrl:Hsa.HRAS-mCherry) s916 , which was absent from this region (Fig. 2g-g’), suggesting labelling of these additional vessels by EGFP likely explains increases in the trunk vascular network observed in this transgenic background. Furthermore, similar to the brain vasculature, vessel diameters were significantly lower when measured using Tg(kdrl:EGFP) s843 (Fig. 2k-l), indicating that diameter measurements are particularly sensitive to the subcellular localisation of fluorescent transgenes. These findings indicate that the selection of transgenes is an important parameter that can influence vascular measurements. To examine the versatility of VISTA-Z across developmental stages, we analysed embryos at 5 dpf when the zebrafish vasculature has undergone further remodelling and maturation [12]. At 5 dpf, both brain and trunk vascular networks displayed increased complexity compared to 3 dpf (Fig. S2a and S2g). No significant differences were observed in normalised network length (Fig. S2b and S2h), vessel density (Fig. S2c and S2i) and number of branchpoints (Fig. S2d and S2j) at this stage, with either transgene in both brain and trunk regions. However, vessel diameter was reduced in Tg(kdrl:EGFP) s843 compared to Tg(kdrl:Hsa.HRAS-mCherry) s916 (Fig. S2e-f and S2k-l). Together, these results establish the robustness of our pipeline and identify transgene-specific effects that can influence vessel measurements. We implemented ROI-based quantification in VISTA-Z to compare different vascular regions within the brain at 3 dpf. We first validated our ROI selection by assessing left–right symmetry in Tg(kdrl:EGFP) s843 ; Tg(kdrl:Hsa.HRAS-mCherry) s916 double transgenic zebrafish embryos (Fig. S3a). There were no significant differences between hemispheres in all quantified vessel metrics, including overall network length, vessel density, or the number of branchpoints (Fig. S3b-e), consistent with the high degree of brain vascular symmetry reported at this developmental stage [10]. No significant differences in the vasculature were observed between the midbrain and hindbrain (Fig. S3f-j), with the exception that hindbrain vessels showed fewer branchpoints and smaller diameters than midbrain vessels in Tg(kdrl:Hsa.HRAS-mCherry) s916 embryos (Fig. S3i and S3j). These differences likely reflect intrinsic biological variation in vessel architecture between these regions. The hindbrain contains larger vessels, such as the basilar artery (BA), and primordial hindbrain channels (PHBCs), which reduce branching complexity, and thinner segments, such as the central arteries (CtAs), which may influence diameter measurements [12]. Consistent with our previous results, vessel diameters were reduced when measured using Tg(kdrl:EGFP) s843 compared to Tg(kdrl:Hsa.HRAS-mCherry) s916 (Fig. S3j). Taken together, these results demonstrate the robustness of VISTA-Z in extracting quantitative vascular data from specific regions of the embryo. Quantification of zebrafish vascular network expansion Vascular networks increase in complexity as they expand during development [12]. Thus, we next compared different developmental stages to determine whether our pipeline could identify this increase in complexity. We used VISTA-Z to quantify development of the vascular network within the brain and trunk between 3 and 5 dpf. In the brain (Fig. 3a-b), network length (Fig. 3c) and vessel density (Fig. 3d) increased progressively. Branchpoints increased between 3 and 4 dpf but then plateaued (Fig. 3e), indicating that most brain vessels form by 4 dpf, with subsequent growth driven by elongation and refinement rather than new branching, in agreement with previous studies [20]. Vessel diameter remained unchanged across timepoints (Fig. 3f), suggesting that higher vascular density reflects network expansion rather than vessel widening. Similar trends were observed in the trunk vasculature (Fig. 3g-h). From 3 to 5 dpf, normalised network length (Fig. 3i) and vessel density (Fig. 3j) increased due to ongoing angiogenesis and remodelling. Unlike the developing brain, branchpoints increased at each stage between 3 and 5 dpf (Fig. 3k), with sustained trunk angiogenesis and vascular remodelling. As in the developing brain, vessel diameter within the trunk showed no significant variation (Fig. 3l). These findings confirm VISTA-Z’s ability to detect developmental changes in vascular architecture during development and establish a reference for future studies. VISTA-Z detects gross vascular abnormalities in the brain in foxc1a mutants We next tested the ability of VISTA-Z to characterise vascular defects in mutant zebrafish embryos. Forkhead box C1 (FOXC1) is a transcription factor which regulate vascular development [21], with mutations in humans linked to cerebral small vessel disease and stroke [22]. We employed a foxc1a sh356 mutant with previously reported loss of central arteries within the developing brain [23]. Quantitative analysis of confocal imaging of foxc1a sh356 ; Tg(kdrl:EGFP) s843 zebrafish embryos (Fig. 4a) indicated reduced normalised network length (Fig. 4b) and vessel density at 3 dpf (Fig. 4c), reflecting loss of brain vascular segments (Fig. 4a) [23–26]. Branchpoints were also reduced, though with higher variability (Fig. 4d). Notably, vessel diameter increased in foxc1a sh356 mutants (Fig. 4e), suggesting that brain network loss in this mutant may alter blood flow dynamics, influencing vessel calibre. We used the VISTA-Z ROI tool to segment the brain vasculature into midbrain and hindbrain regions using the middle cerebral vein (MCeV) as a boundary (Fig. 4f). Regional analysis highlighted reduced network length (Fig. 4g), vessel density (Fig. 4h) and branchpoints (Fig. 4i) in both midbrain and hindbrain regions of foxc1a mutants. While vessel diameter showed no significant differences between regions (Fig. 4j), variability was markedly higher than in the overall brain vasculature analysis (Fig. 4e and 4j). This pattern likely reflects the absence of arterial segments such as CtAs, BA, and posterior and middle mesencephalic central arteries (PMCtA and MMCtA) in foxc1a mutants, whereas venous structures, including MCeVs, posterior cerebral veins (PCeVs), dorsal longitudinal vein (DLV), mesencephalic veins (MSVs), and anterior cerebral vein (ACeV) appeared morphologically normal (Fig. 4f). These findings indicate a preferential requirement and previously unrealised role for foxc1a in midbrain and hindbrain arterial development. Overall, VISTA-Z is capable of reliably detecting regional vascular abnormalities. VISTA-Z detects global vascular patterning defects in kdrl mutants kdrl encodes a zebrafish orthologue of mammalian VEGFR2/KDR (Fig. S4), a receptor tyrosine kinase essential for endothelial cell proliferation, migration, and angiogenic sprouting [27,28]. Quantitative analysis of kdrl qmc313 ; Tg(kdrl:EGFP) s843 zebrafish embryos (Fig. 5a) showed significant reductions in normalised network length (Fig. 5b), vessel density (Fig. 5c), and branchpoints at 3 dpf (Fig. 5d), consistent with extensive cerebrovascular loss in previously reported kdrl mutants [28,29]. Vessel diameter was increased (Fig. 5e), possibly reflecting compensatory haemodynamic changes in kdrl mutants. We segmented the midbrain and hindbrain vasculature in kdrl mutants and compared this to controls (Fig. S5a). Regional quantification revealed reduced network length (Fig. S5b), vessel density (Fig. S5c), and branchpoints (Fig. S5d) in both brain regions. Vessel diameter was not significantly different but trended toward larger diameter in mutants (Fig. S5e). These findings indicate that both arteries and veins were abnormal in kdrl mutants, underscoring widespread cerebrovascular disruption. In the trunk (Fig. 5f), kdrl mutants displayed reduced normalised network length (Fig. 5g), vessel density (Fig. 5h), and branchpoints (Fig. 5i) similar to observations in the brain vasculature. In addition, kdrl mutants displayed increased trunk vessel diameter (Fig. 5j). To map spatial vascular patterning, we applied the VISTA-Z ROI tool to segment the trunk into anterior (ISVs 1-4), medial (ISVs 5-25), and posterior (ISVs 26-30) regions based on the stereotypical position of ISVs in wild type embryos at 3 dpf (Fig. 3h) [28]. The anterior and medial trunk regions showed the most severe phenotype, with the medial trunk displaying near-complete ISV loss (Fig. S6a and S6b), consistent with previous studies [30,31]. In contrast, posterior ISVs formed and connected relatively normally (Fig. S6b), suggesting reduced dependence on kdrl signalling during angiogenesis in the posterior trunk. Quantitative analysis indicated a significant reduction in network length (Fig. S6d) and branchpoints (Fig. S6f) in the medial trunk, with a trend towards reduction in the anterior region. Vessel density was markedly reduced in the anterior and medial trunk, consistent with ISV loss, whereas the posterior trunk exhibited a more variable and less pronounced reduction in ISVs (Fig. S6e). Vessel diameter was not significantly different when assessed by brain region (Fig. S6g). Collectively, these results demonstrate that VISTA-Z enables detection of regional vascular changes in anatomically and morphologically distinct territories. VISTA-Z reliably quantifies hyper-vascularisation in plxnd1 crispants After characterising mutants with reduced vascular networks, we applied VISTA-Z to analyse plxnd1 crispants which display elevated vascular network formation (Fig. 6). plxnd1 mutation causes excessive angiogenic sprouting, aberrant vessel branching, and disorganised vascular patterning (Fig. 6a) [32,33]. We generated mosaic plxnd1 crispants via CRISPR–Cas9 in Tg(kdrl:EGFP) s843 embryos. Quantification showed significant increases in normalised network length (Fig. 6b), vessel density (Fig. 6c), and branchpoints (Fig. 6d), indicating excessive vessel formation. Unlike foxc1a and kdrl mutants, vessel diameters in plxnd1 crispants remained unchanged in the developing brain (Fig. 6e). To characterise the spatial distribution of vascular overgrowth, we employed VISTA-Z to segment the brain into midbrain and hindbrain regions using the MCeVs as a boundary (Fig. S7a). Quantitative analysis at 3 dpf indicated significant increases in normalised network length (Fig. S7b), vessel density (Fig. S7c), and branchpoints in both regions (Fig. S7d), while vessel diameter was not significantly different between plxnd1 crispants and controls (Fig. S7e). The brain vascular overgrowth observed in plxnd1 crispants is likely due to hyper-angiogenesis of the CtAs, MMCtA and PMCtA and aberrant anastomosis of the PCeV and ACeV (Fig. S7a). These findings suggest that plxnd1 disruption drives ectopic angiogenesis across the brain vasculature. Throughout the trunk (Fig. 6f), plxnd1 crispants showed significant increases in network length (Fig. 6g), vessel density (Fig. 6h), and branchpoints at 3 dpf (Fig. 6i), consistent with pronounced hyper-vascularisation. Vessel diameter decreased (Fig. 6j), likely reflecting immature and unlumenised vessels. Applying VISTA-Z to trunk segmentation (Fig. S8a), all regions (anterior, medial and posterior) showed increased network length (Fig. S8b) and vessel density (Fig. S8c) at 3 dpf, with branchpoints elevated in the anterior and medial trunk but not in the posterior trunk (Fig. S8d). Vessel diameters were not significantly different throughout the trunk in plxnd1 crispants (Fig. S8e). Notably, ventral vascular beds such as the SIVP expanded into normally avascular territories (Fig. S8f). The SIVP of plxnd1 crispants exhibited increased network length (Fig. S8g), vessel density (Fig. S8h), and branchpoints (Fig. S8i), without significant changes in vessel diameter (Fig. S8j). Collectively, these results indicate VISTA-Z detects both vascular expansion and reduction, underscoring its utility for characterising diverse vascular phenotypes. Discussion Quantitative analysis of vascular architecture is fundamental for understanding angiogenesis and vascular pathology, yet current zebrafish imaging workflows remain constrained by manual segmentation and limited scalability. Existing automated tools have improved accessibility but often lack robustness across diverse imaging conditions, anatomical regions and developmental stages [10,11]. VISTA-Z addresses these gaps by introducing a fully integrated pipeline that combines unsupervised vessel segmentation with segment-level labelling, facilitating optional manual curation, branchpoint refinement and systematic filtering and normalisation of vessel metrics, advancing beyond prior frameworks that rely on extensive manual correction. Validation across double-transgenic lines, multiple developmental stages, and different anatomical regions in distinct gain- and loss-of-function zebrafish mutants confirms the robustness of VISTA-Z. We first analysed the embryonic development of the vasculature in transgenic embryos between 3 and 5 dpf, enabling detection of vascular changes during development. These results demonstrate VISTA-Z’s capacity to identify significant changes in vascular architecture across developmental stages and anatomical regions, underscoring its applicability for pre-clinical zebrafish models. Automated pipelines have transformed vascular image analysis, yet many existing approaches struggle with segmentation accuracy. This often results from misclassifying non-vascular regions or from poor performance in areas with low signal-to-noise contrast [10]. To overcome these limitations, VISTA-Z employs a novel two-step segmentation strategy. First, an unsupervised Meijering-based filter is applied for automated vessel detection [14]. Next, individual segment labelling enables targeted artefact removal, generating high-quality vessel masks with minimal user intervention. Manual curation continues to play a vital role in vascular analysis, especially with complex embryonic datasets, where fluorescent artefacts or incomplete vessel lumenisation can interfere with segmentation [8]. Our two-step approach offers a robust and scalable solution in the absence of established standards for vessel segmentation and vascular analysis [8,11]. The inclusion of segment-level labelling for artefact removal represents a novel approach to improve segmentation accuracy without sacrificing automation, addressing a persistent challenge in embryonic datasets characterised by low signal-to-noise ratios. A key innovation of VISTA-Z lies in its ability to standardise vascular measurements across heterogeneous datasets, enabling reproducible comparisons between experiments and laboratories. Our branchpoint refinement strategy addresses the overestimation of segment connections that arises when 3D vascular structures are projected onto 2D images using maximum intensity projections [11]. Although full 3D reconstruction is ideal, it is often constrained by low signal and imaging artefacts [10,16]. Therefore, we implemented distance-based filtering to reduce the false detection of branchpoints. Moreover, metric normalisation by z-depth and conversion into spatial units standardise measurements across the diverse imaging modalities used in the field. This ensures that differences reflect biological variation rather than technical inconsistencies. Previous frameworks often lacked these corrections, introducing variability and limiting biological interpretability [11,15]. Our refinements mitigate these limitations, improving cross-study reproducibility and enabling improved vascular comparisons. Validating image analysis frameworks across multiple fluorescent reporters and anatomical regions is crucial to ensure their robustness [16,34]. Contrast-to-noise ratios have been shown to vary between zebrafish transgenic lines but remain relatively stable across developmental stages and vessel sizes [8]. Using double transgenic lines provides complementary vascular information, improving segmentation accuracy and mitigating artefacts associated with single-reporter analysis [16]. The use of multiple reporters also addresses common issues, such as variation in promoter activity and fluorophore localisation, which can impact vessel measurements and lead to misinterpretation of results [17,35]. Notably, our data show reporter-specific differences in vessel diameter measurements in both the brain and trunk vasculature across different developmental stages. This underscores the importance of considering reporter cellular localisation in experiments, which can influence vessel metrics. The ability of VISTA-Z to detect significant differences in vessel diameter between membrane or cytoplasmic-localised endothelial transgenic lines highlights its sensitivity and potential utility in identifying subtle vascular changes. Validation across anatomically distinct vascular territories further demonstrates the robustness and utility of VISTA-Z to characterise vascular regions which differ in complexity and morphology [12]. Defining how morphometric parameters such as network length, branchpoint density, and vessel diameter change during development is essential to distinguish pathological alterations from normal developmental variation [36]. We observed linear increases in network length and branchpoint density, without notable changes in vessel diameter, indicating that vascular expansion in zebrafish between 3 and 5 dpf primarily reflects elongation and branching rather than lumen widening, consistent with angiogenic remodelling [37]. Applying VISTA-Z to study the vasculature of genetic mutants highlights its ability to detect diverse angiogenic perturbations. Our results reinforce the role of foxc1a in arterial specification and angiogenesis [21,23]. FOXC1 is a transcription factor implicated in vascular development and endothelial cell migration [21,38]. While previous studies, including our own, primarily reported CtA defects in the hindbrain of foxc1a zebrafish mutant embryos [23,26,39], VISTA-Z analysis reveals that the midbrain vasculature is also compromised, indicating that foxc1a is required more widely within brain vascular patterning. VISTA-Z analysis also uncovered an increase in vessel diameter in foxc1a mutants, which we were previously unaware of, likely reflecting compensatory haemodynamic adjustments due to reduced network complexity. Comparable compensation occurs in alk1 zebrafish mutants, in which arteriovenous malformations and vessel dilation occur in response to altered shear stress sensing [40,41]. Reduced vascular resistance and abnormal flow in alk1 mutants trigger adaptive increases in vessel diameter [40,42], suggesting similar mechanisms may function in foxc1a mutants. Kdrl is a membrane receptor essential for angiogenic sprouting, endothelial proliferation, migration, and survival, which mediates VEGF signalling [27,30]. Our kdrl qmc313 mutants display severe brain vascular defects, including significant midbrain network disruption, consistent with previously reported kdrl mutants [27–29]. While kdrl is dispensable for primary vasculogenesis [27,28,43], it is essential to promote elaboration of complex vascular networks via angiogenesis. We quantified ISV defects along the trunk of kdrl mutant embryos, revealing a pronounced loss in the medial trunk (ISVs 4–25), while posterior ISVs (25–30) were largely preserved, indicating region-specific requirements for kdrl during trunk angiogenesis. Validation of VISTA-Z using plxnd1 crispants demonstrates its utility in characterising hyper-angiogenic phenotypes. Plxnd1, a transmembrane receptor for class 3 Semaphorins, is a crucial negative regulator of angiogenic sprouting by both antagonising VEGF signalling and by providing repulsive cues to migrating endothelial tip cells [32,33]. While previous studies focused on its role in trunk vasculature development [32,33,44], our analysis shows that plxnd1 also limits brain angiogenesis. In the trunk, plxnd1 crispants exhibit pronounced vascular overgrowth, along with reduced vessel diameters, likely due to the presence of immature and unlumenised angiogenic sprouts, which lack functional lumens and only expand in diameter as they mature and experience blood flow [45]. VISTA-Z provides a robust, automated framework for zebrafish vascular analysis, combining accurate segmentation with standardised metric extraction and regional quantification. By enabling reproducible phenotyping across developmental stages and anatomical regions, this pipeline addresses key limitations of existing tools and offers a scalable solution for high-throughput vascular studies. Collectively, these findings highlight VISTA-Z's ability to resolve spatially distinct vascular phenotypes and to uncover mechanisms that may be overlooked by manual approaches. Materials and Methods Zebrafish strains and husbandry Zebrafish husbandry and experimental procedures were approved by the University of Nottingham Animal Welfare and Ethical Review Body and conducted under UK Home Office Licence 4626831. All methods were performed in accordance with institutional and national animal welfare guidelines, Directive 2010/63/EU of the European Parliament, and ARRIVE guidelines on the protection of animals used for scientific purposes. The following published lines were employed: Tg(kdrl:EGFP) s843 [46], Tg(kdrl:Hsa.HRAS-mCherry) s916 [17], Tg(fli1a:nls-mCherry) sh550 [47] and foxc1a sh356 mutants [23], and nacre w2 mutants [48] to facilitate confocal imaging within these transgenic and mutant backgrounds. Adult zebrafish were housed following standard husbandry protocols [49]. Embryos were obtained by controlled pair-mating and maintained in E3 buffer (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, and 0.33 mM MgSO 4 ). All embryo imaging was performed under anaesthesia using tricaine at 160 mg/mL, and euthanasia was performed by extended incubation in this solution until permanent cessation of circulation. Generation of a novel kdrl mutant allele CRISPR Cas9 mutagenesis was performed as described [50]. Briefly, a predesigned Alt-R™ CRISPR RNA (crRNA) (IDT) targeting kdrl (ENSDARG00000105215) at 5’-GGGACCGTGAAACCTTTTTTTGG-3’ was annealed with trans-activating CRISPR RNA (tracrRNA) (IDT) to form single guide RNA (sgRNA), which was complexed with Engen ® Spy Cas9 NLS protein (NEB) and injected into one-cell stage embryos [50]. sgRNA efficiency was assessed by high-resolution melt analysis using the following primers: forward: 5’-TGGCAGATGATGTTAAAAGAGGG-3’ reverse: 5’-CACCCTGGTCAAGCATGTAA-3’. G0 CRISPR Cas9-injected adults were incrossed, and progeny were genotyped for kdrl qmc313 allele as described [51]. kdrl qmc313 is a 2bp deletion within exon 5, which destroys a BslI restriction site and is predicted to prematurely truncate before the conserved transmembrane and intracellular tyrosine kinase domains [27]. Generation of plxnd1 G0 crispants CRISPR Cas9 mutagenesis was carried out as described above. Guide RNAs (crRNAs) were designed using the Alt-R™ CRISPR-Cas9 system (IDT) to target exon 1 of plxnd1 (ENSDARG00000086057). The selected target sequence was 5’-GGTGCTCGCGTTCTCGTGG-3’ [52]. crRNA and tracrRNA were annealed and complexed with Engen ® Spy Cas9 NLS protein (NEB) before injection into one-cell stage embryos [50]. Crispants were identified based on the previously described vascular phenotype, characterised by excessive sprouting extending beyond intersegmental vessel (ISV)-somite boundaries [32,33,52]. Confocal imaging For live confocal imaging, embryos were anaesthetised in tricaine, embedded in 0.7% low-melting point agarose (Fisher Scientific) in E3 buffer and mounted on 27 mm Nunc™ glass bottom dishes (ThermoFisher). Imaging was performed on a Zeiss LSM880 confocal microscope with Zen Black software (Zeiss) using 10x (NA 0.3) and 20x (NA 0.8) air objectives. All images are 2D maximum intensity projections of 3D z-stacks acquired with a 5 mm step size. Data were acquired using line-sequential scanning mode at an XY resolution of 0.83 mm/pixel, and images were captured as 16-bit data (1024x1024). Image processing and analysis were performed using Python (version 3.10.11); code is available on GitHub: https://github.com/fishyvessels/VISTA-Z Image pre-processing and segmentation Maximum intensity projections were obtained using the CziFile library (version 2019.7.2.1) [53]. For two-colour acquisitions, channels were processed independently before analysis. Images were pre-processed by normalising to an 8-bit format and enhancing contrast using Contrast Limited Adaptive Histogram Equalisation (CLAHE; clip limit = 2.0, tile grid size = 4×4) implemented in OpenCV (version 4.11.0) [54] to improve vessel visibility across heterogeneous backgrounds. Initial binary masks were generated using Otsu thresholding [55]. Segmentation of vascular structures was performed using Meijering filtering (segmentation threshold: 10, sigma range: 3-8 pixels, step size: 1 pixel), followed by morphological post-processing to remove small artefacts (<50 pixels) and fill holes (<200 pixels). Our pipeline increases functionality present in the VesselMetrics Python package [11], by introducing custom modules for segment labelling and user-based quality control. Vessel regions were automatically labelled using connected component analysis and visualised using a pseudo-colour palette. Segmentation results were manually curated through an interactive quality control process where non-vascular artefacts were identified by visual inspection and removed from subsequent analysis. These additions improve segmentation accuracy and downstream analysis. Automated image-based skeleton, vessel metric analysis and outlier filtering Quantitative analysis of the zebrafish vasculature was performed using a combination of core functions from the VesselMetrics Python package [11] and our novel modules, including refined branchpoint filtering and metric normalisation to improve biological relevance and scalability in zebrafish embryo datasets. Briefly, segmented vessel masks were converted into simplified centreline representations using morphological thinning to reduce vessels to single-pixel-wide paths. Branchpoints were identified as skeleton pixels with more than three connected neighbours in a 3x3 connectivity matrix. To remove detection artefacts from clustered branchpoints, we implemented a distance-based filtering approach. Pairwise Euclidean distances were computed between all detected branchpoints, and only the first point within a minimum-distance threshold (mean vessel length ≈ 30 pixels) was retained. Vessel segments were defined as skeleton regions between branchpoints and endpoints, identified through connected component analysis. Vessel diameters were estimated by calculating the maximum inscribed circle diameter at each skeleton pixel using distance transform methods. Individual vascular segment lengths were calculated as the sum of skeleton pixels within each connected segment. Network-level metrics included vascular length (total skeleton length), vessel density (vessel pixel density computed on tiled regions sized to represent 100 µm 2 windows), and branchpoint density (number of branchpoints per unit area) [11]. All measurements were converted from pixels to physical units (microns) using an empirically determined pixel-to-micron scale factor of 1.2 pixels/µm (using Zeiss LSM880 confocal microscope with a 10x (NA 0.3) air objective). Outlier removal was applied to diameter and vessel length distributions using a median absolute deviation-based filtering (MAD) [56,57], with asymmetric thresholds (lower = 1 × MAD, upper = 3 × MAD). Results were exported as Excel files for downstream data visualisation and statistical analysis. Statistical analysis Statistical analysis was performed using GraphPad Prism (10.6.1). Normality was assessed using the Shapiro-Wilk test. Sample sizes (biological replicates from ≥2 breeding pairs) are indicated in figure legends. Data are shown as box-and-whisker plots (median, interquartile range). Two group comparisons used paired or unpaired Student’s t-test (normally distributed data) and Mann–Whitney U test (non-normally distributed data). For ≥3 groups, one- or two-way ANOVA with Tukey’s test (normal) or Kruskal–Wallis with FDR correction (non-normal) was applied. All values were considered significant with p-values ≤ 0.05, (ns, p > 0.05; * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001). Samples with poor image quality were excluded from analysis. Declarations Data availability statement Code is available via GitHub (https://github.com/fishyvessels/VISTA-Z). Other data available upon reasonable request from the corresponding author. Funding This work was supported by MRC MR/X008215/1 to RNW. Acknowledgements We are grateful for confocal imaging support from the University of Nottingham School of Life Sciences Imaging (SLIM) facility, expert zebrafish husbandry assistance from Charlotte Moore and Harry Gilbert. Author Contribution RNW and IRP conceived the project. IRP, JR and RNW designed experiments. IRP and JR performed experiments. IRP and RNW analysed data. IRP and RNW wrote the manuscript and all authors reviewed the final manuscript. Additional information - Declaration of competing interest The authors declare no competing of interest. References Alexander Y, Osto E, Schmidt-Trucksäss A, Shechter M, Trifunovic D, Duncker DJ, et al. Endothelial function in cardiovascular medicine: a consensus paper of the European Society of Cardiology Working Groups on Atherosclerosis and Vascular Biology, Aorta and Peripheral Vascular Diseases, Coronary Pathophysiology and Microcirculation, and Thrombosis. Cardiovasc Res 2021;117:29–42. https://doi.org/10.1093/CVR/CVAA085. Ghaffari S, Leask RL, Jones EAV. Simultaneous imaging of blood flow dynamics and vascular remodelling during development. Development (Cambridge) 2015;142:4158–67. https://journals.biologists.com/dev/article/142/23/4158/47121. Rödel CJ, Otten C, Donat S, Lourenço M, Fischer D, Kuropka B, et al. Blood Flow Suppresses Vascular Anomalies in a Zebrafish Model of Cerebral Cavernous Malformations. Circ Res 2019;125:e43–54. https://doi.org/10.1161/CIRCRESAHA.119.315076. Wilkinson RN, Van Eeden FJM. The zebrafish as a model of vascular development and disease. Prog Mol Biol Transl Sci 2014;124:93–122. https://doi.org/10.1016/B978-0-12-386930-2.00005-7. Howe K, Clark MD, Torroja CF, Torrance J, Berthelot C, Muffato M, et al. The zebrafish reference genome sequence and its relationship to the human genome. Nature 2013;496:498–503. https://doi.org/10.1038/nature12111. Gut P, Reischauer S, Stainier DYR, Arnaout R. Little fish, big data: Zebrafish as a model for cardiovascular and metabolic disease. Physiol Rev 2017;97:889–938. https://doi.org/10.1152/PHYSREV.00038.2016. Chico TJA, Ingham PW, Crossman DC. Modeling Cardiovascular Disease in the Zebrafish. Trends Cardiovasc Med 2008;18:150–155. https://doi.org/10.1016/J.TCM.2008.04.002. Kugler E, Plant K, Chico T, Armitage P. Enhancement and Segmentation Workflow for the Developing Zebrafish Vasculature. J Imaging 2019;5:14-27. https://doi.org/10.3390/JIMAGING5010014. Dash S, Verma S, Kavita, Khan MS, Wozniak M, Shafi J, et al. A Hybrid Method to Enhance Thick and Thin Vessels for Blood Vessel Segmentation. Diagnostics 2021;11:2017-2036. https://doi.org/10.3390/DIAGNOSTICS11112017. Kugler E, Frost J, Silva V, Plant K, Chhabria K, Chico TJA, et al. Zebrafish vascular quantification: A tool for quantification of three-dimensional zebrafish cerebrovascular architecture by automated image analysis. Development (Cambridge) 2022;149: dev199720-dev199735. https://journals.biologists.com/dev/article/149/3/dev199720/274364. McGarry SD, Adjekukor C, Ahuja S, Greysson-Wong J, Vien I, Rinker KD, et al. Vessel Metrics: A software tool for automated analysis of vascular structure in confocal imaging. Microvasc Res 2024;151:104610–22. https://doi.org/10.1016/j.mvr.2023.104610. Isogai S, Horiguchi M, Weinstein BM. The Vascular Anatomy of the Developing Zebrafish: An Atlas of Embryonic and Early Larval Development. Dev Biol 2001;230:278–301. https://doi.org/10.1006/DBIO.2000.9995. da Rocha DA, Barbosa ABL, Guimarães DS, Gregório LM, Gomes LHN, da Silva Amorim L, et al. An unsupervised approach to improve contrast and segmentation of blood vessels in retinal images using CLAHE, 2D Gabor wavelet, and morphological operations. Research on Biomedical Engineering 2020;36:67–75. https://doi.org/10.1007/S42600-019-00032-Z. Lamy J, Merveille O, Kerautret B, Passat N, Vacavant A. Vesselness filters: A survey with benchmarks applied to liver imaging. Int Conf Pattern Recognit 2020:3528–3535. https://doi.org/10.1109/ICPR48806.2021.9412362. McDowell KP, Berthiaume AA, Tieu T, Hartmann DA, Shih AY. VasoMetrics: unbiased spatiotemporal analysis of microvascular diameter in multi-photon imaging applications. Quant Imaging Med Surg 2021;11:969-983. https://doi.org/10.21037/QIMS-20-920. Daetwyler S, Gü Nther U, Modes CD, Harrington K, Huisken J. Multi-sample SPIM image acquisition, processing and analysis of vascular growth in zebrafish. Development (Cambridge) 2019,6:146–177. https://doi.org/10.1242/dev.173757. Chi NC, Shaw RM, De Val S, Kang G, Jan LY, Black BL, et al. Foxn4 directly regulates tbx2b expression and atrioventricular canal formation. Genes Dev 2008;22:734–739. https://doi.org/10.1101/GAD.1629408. Wild R, Klems A, Takamiya M, Hayashi Y, Strähle U, Ando K, et al. Neuronal sFlt1 and Vegfaa determine venous sprouting and spinal cord vascularization. Nat Commun 2017;8:1–17. https://doi.org/10.1038/ncomms13991. Schevenels G, Cabochette P, America M, Vandenborne A, De Grande L, Guenther S, et al. A brain-specific angiogenic mechanism enabled by tip cell specialization. Nature 2024;628:1–9. https://doi.org/10.1038/s41586-024-07283-6. Fujita M, Cha YR, Pham VN, Sakurai A, Roman BL, Silvio Gutkind J, et al. Assembly and patterning of the vascular network of the vertebrate hindbrain. Development 2011;138:1705–1715. https://doi.org/10.1242/DEV.058776. Kume T. The Cooperative Roles of Foxc1 and Foxc2 in Cardiovascular Development. Adv Exp Med Biol 2009;665:63–77. https://doi.org/10.1007/978-1-4419-1599-3_5. French CR, Waskiewicz AJ, Lehmann OJ. Mutation of FOXC1 and PITX2 induces cerebral small-vessel disease. J Clin Invest 2014;124:4881-4887. https://doi.org/10.1172/JCI75109. Jiang Z, Evans T, Savage AM, Loose M, Chico TJA, Van Eeden FJM, et al. foxc1a and foxc1b differentially regulate angiogenesis from arteries and veins by modulating Vascular Endothelial Growth Factor signalling. BioRxiv 2018:417931. https://doi.org/10.1101/417931. Skarie JM, Link BA. FoxC1 is essential for vascular basement membrane integrity and hyaloid vessel morphogenesis. Invest Ophthalmol Vis Sci 2009;50:5026-5034. https://doi.org/10.1167/IOVS.09-3447. Whitesell TR, Chrystal PW, Ryu JR, Munsie N, Grosse A, French CR, et al. foxc1 is required for embryonic head vascular smooth muscle differentiation in zebrafish. Dev Biol 2019;453:34–47. https://doi.org/10.1016/J.YDBIO.2019.06.005. Duan X, Shi Y, Zhao S, Yao L, Sheng J, Liu D. Foxc1a regulates zebrafish vascular integrity and brain vascular development through targeting amotl2a and ctnnb1. Microvasc Res 2022;143:104400. https://doi.org/10.1016/J.MVR.2022.104400. Habeck H, Odenthal J, Walderich B, Maischein HM, Bebber van F, Busch-Nentwich E, et al. Analysis of a Zebrafish VEGF Receptor Mutant Reveals Specific Disruption of Angiogenesis. Current Biology 2002;12:1405–1412. https://doi.org/10.1016/S0960-9822(02)01044-8. Bussmann J, Wolfe SA, Siekmann AF. Arterial-venous network formation during brain vascularization involves hemodynamic regulation of chemokine signaling. Development (Cambridge) 2011;138:1717–1726. https://doi.org/10.1242/DEV.059881. Fetsko AR, Sebo DJ, Taylor MR. Brain endothelial cells acquire blood-brain barrier properties in the absence of Vegf-dependent CNS angiogenesis. Dev Biol 2023;494:46–59. https://doi.org/10.1016/J.YDBIO.2022.11.007. Covassin LD, Villefranc JA, Kacergis MC, Weinstein BM, Lawson ND. Distinct genetic interactions between multiple Vegf receptors are required for development of different blood vessel types in zebrafish. Proc Natl Acad Sci U S A 2006;103:6554-6559. https://doi.org/10.1073/PNAS.0506886103. Covassin LD, Siekmann AF, Kacergis MC, Laver E, Moore JC, Villefranc JA, et al. A genetic screen for vascular mutants in zebrafish reveals dynamic roles for Vegf/Plcg1 signaling during artery development. Dev Biol 2009;329:212-226. https://doi.org/10.1016/J.YDBIO.2009.02.031. Torres-Vázquez J, Gitler AD, Fraser SD, Berk JD, Pham VN, Fishman MC, et al. Semaphorin-plexin signaling guides patterning of the developing vasculature. Dev Cell 2004;7:117–123. https://doi.org/10.1016/j.devcel.2004.06.008. Zygmunt T, Gay CM, Blondelle J, Singh MK, Flaherty KMC, Means PC, et al. Semaphorin-PlexinD1 Signaling Limits Angiogenic Potential via the VEGF Decoy Receptor sFlt1. Dev Cell 2011;21:301–314. https://doi.org/10.1016/j.devcel.2011.06.033. Page DJ, Thuret R, Venkatraman L, Takahashi T, Bentley K, Herbert SP. Positive Feedback Defines the Timing, Magnitude, and Robustness of Angiogenesis. Cell Rep 2019;27:3139-3151. https://doi.org/10.1016/J.CELREP.2019.05.052. Victoria AS, Roy B, Victoria LH. Development of an ImageJ-based method for analysing the developing zebrafish vasculature. Vasc Cell 2017;9:21-40. https://doi.org/10.24238/13221-9-1-172. Martin-Valiente E, Du Y, Goemans C, America M, Zindy E, Adam M, et al. Reversal of cerebrovascular anomalies in a zebrafish model of vein of Galen aneurysm. Nat Cardiovasc Res 2025;4:773–789. https://doi.org/10.1038/s44161-025-00659-5. Gore A V., Monzo K, Cha YR, Pan W, Weinstein BM. Vascular Development in the Zebrafish. Cold Spring Harb Perspect Med 2012;2:a006684. https://doi.org/10.1101/CSHPERSPECT.A006684. Hayashi H, Kume T. Forkhead transcription factors regulate expression of the chemokine receptor CXCR4 in endothelial cells and CXCL12-induced cell migration. Biochem Biophys Res Commun 2008;367:584–589. https://doi.org/10.1016/J.BBRC.2007.12.183. Topczewska JM, Topczewski J, Solnica-Krezel L, Hogan BLM. Sequence and expression of zebrafish foxc1a and foxc1b, encoding conserved forkhead/winged helix transcription factors. Mech Dev 2001;100:343–347. https://doi.org/10.1016/S0925-4773(00)00534-7. Corti P, Young S, Chen CY, Patrick MJ, Rochon ER, Pekkan K, et al. Interaction between alk1 and blood flow in the development of arteriovenous malformations. Development (Cambridge) 2011;138:1573–1582. https://doi.org/10.1242/DEV.060467. Diwan Z, Kang J, Tsztoo E, Siekmann AF. Alk1/Endoglin signaling restricts vein cell size increases in response to hemodynamic cues. Angiogenesis 2024;28:20-35. https://doi.org/10.1007/S10456-024-09955-3. ochon ER, Menon PG, Roman BL. Alk1 controls arterial endothelial cell migration in lumenized vessels. Development (Cambridge) 2016;143:2593–2602. https://journals.biologists.com/dev/article/143/14/2593/47581. Bahary N, Goishi K, Stuckenholz C, Weber G, LeBlanc J, Schafer CA, et al. Duplicate VegfA genes and orthologues of the KDR receptor tyrosine kinase family mediate vascular development in the zebrafish. Blood 2007;110:3627–3636. https://doi.org/10.1182/BLOOD-2006-04-016378. Childs S, Chen JN, Garrity DM, Fishman MC. Patterning of angiogenesis in the zebrafish embryo. Development (Cambridge) 2002;129:973–982. https://doi.org/10.1242/DEV.129.4.973. Gebala V, Collins R, Geudens I, Phng LK, Gerhardt H. Blood flow drives lumen formation by inverse membrane blebbing during angiogenesis in vivo. Nat Cell Biol 2016;18:443–450. https://doi.org/10.1038/ncb3320. Beis D, Bartman T, Jin SW, Scott IC, D’Amico LA, Ober EA, et al. Genetic and cellular analyses of zebrafish atrioventricular cushion and valve development. Development (Cambridge) 2005;132:4193–4204. https://doi.org/10.1242/DEV.01970. Chen Y, Savage AM, Jiang Z, Kim HR, Evans PC, Wilkinson RN. Blood flow coordinates collective endothelial cell migration during vascular plexus formation and promotes angiogenic sprout regression via vegfr3/flt4. BioRxiv 2025. https://doi.org/10.1101/2021.07.23.453496. Lister JA, Robertson CP, Lepage T, Johnson SL, Raible DW. nacre encodes a zebrafish microphthalmia-related protein that regulates neural-crest-derived pigment cell fate. Development (Cambridge) 1999;126:3757–3767. https://doi.org/10.1242/DEV.126.17.3757. Aleström P, D’Angelo L, Midtlyng PJ, Schorderet DF, Schulte-Merker S, Sohm F, et al. Zebrafish: Housing and husbandry recommendations. Lab Anim 2020;54:213–224. https://doi.org/10.1177/0023677219869037. Kroll F, Powell GT, Ghosh M, Gestri G, Antinucci P, Hearn TJ, et al. A simple and effective f0 knockout method for rapid screening of behaviour and other complex phenotypes. Elife 2021;10:1–34. https://doi.org/10.7554/ELIFE.59683. Wilkinson RN, Elworthy S, Ingham PW, van Eeden FJM. A Method for High-Throughput PCR-Based Genotyping of Larval Zebrafish Tail Biopsies. Biotechniques 2013;55:314–316. https://doi.org/10.2144/000114116. Davis AE, Castranova D, Weinstein BM. Rapid Generation of Pigment Free, Immobile Zebrafish Embryos and Larvae in Any Genetic Background Using CRISPR-Cas9 dgRNPs. Zebrafish 2021;18:242-250. https://doi.org/10.1089/ZEB.2021.0011. Gohlke C. czifile: Read Carl Zeiss image files (CZI). 2013. https://github.com/cgohlke/czifile. Bradski G. The Opencv Library. Doctor Dobb’s Journal of Software Tools, 2000;120;122-125. https://drdobbs.com/ftp/2000/2000_11/opencv.txt. Otsu N. A Threshold Selection Method from Gray-Level Histograms. IEEE Trans Syst Man Cybern 1979;SMC-9:62–66. https://doi.org/10.1109/TSMC.1979.4310076. Rousseeuw PJ, Croux C. Alternatives to the Median Absolute Deviation. J Am Stat Assoc 1993;88:1272-1283. https://doi.org/10.2307/2291267. Leys C, Ley C, Klein O, Bernard P, Licata L. Detecting outliers: Do not use standard deviation around the mean, use absolute deviation around the median. J Exp Soc Psychol 2013;49:764–766. https://doi.org/10.1016/J.JESP.2013.03.013 Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterial.pdf Cite Share Download PDF Status: Published Journal Publication published 01 Apr, 2026 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 03 Feb, 2026 Reviews received at journal 01 Feb, 2026 Reviewers agreed at journal 22 Jan, 2026 Reviews received at journal 20 Jan, 2026 Reviewers agreed at journal 20 Jan, 2026 Reviewers invited by journal 20 Jan, 2026 Editor assigned by journal 20 Jan, 2026 Editor invited by journal 13 Jan, 2026 Submission checks completed at journal 08 Jan, 2026 First submitted to journal 08 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8402098","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":578291108,"identity":"2240498c-a3f7-4d04-8a6f-0b530bb51863","order_by":0,"name":"Ignacio Rodriguez-Pastrana","email":"","orcid":"","institution":"University of Nottingham","correspondingAuthor":false,"prefix":"","firstName":"Ignacio","middleName":"","lastName":"Rodriguez-Pastrana","suffix":""},{"id":578291110,"identity":"3d3d9dc3-4989-418d-b217-380963fd2363","order_by":1,"name":"Joanna Richens","email":"","orcid":"","institution":"University of Nottingham","correspondingAuthor":false,"prefix":"","firstName":"Joanna","middleName":"","lastName":"Richens","suffix":""},{"id":578291111,"identity":"b8642bda-f5ce-427d-8b28-63fcd433e85e","order_by":2,"name":"Robert N. Wilkinson","email":"data:image/png;base64,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","orcid":"","institution":"University of Nottingham","correspondingAuthor":true,"prefix":"","firstName":"Robert","middleName":"N.","lastName":"Wilkinson","suffix":""}],"badges":[],"createdAt":"2025-12-19 07:54:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8402098/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8402098/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-026-43301-5","type":"published","date":"2026-04-01T16:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":100866584,"identity":"81eb9895-01d1-4835-a263-3c52f3da942e","added_by":"auto","created_at":"2026-01-22 08:34:44","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":5370052,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVISTA-Z workflow for zebrafish vascular quantification. a\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eRepresentative confocal images of the brain (left) and trunk (right) vasculature of zebrafish embryos at 3 dpf showing overlay of \u003cem\u003eTg(kdrl:EGFP)\u003c/em\u003e\u003csup\u003e\u003cem\u003es843\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e;Tg(kdrl:Hsa.HRAS-mCherry)\u003c/em\u003e\u003csup\u003e\u003cem\u003es916\u003c/em\u003e\u003c/sup\u003e (top) or individual channels (bottom). \u003cstrong\u003eb\u003c/strong\u003e) Image pre-processing steps including pixel intensity normalisation (1), CLAHE enhancement (2), and Otsu threshold-based masking (3). \u003cstrong\u003eb′\u003c/strong\u003e) Optional ROI selection for detailed vascular analysis. Diagram highlights left/right and midbrain versus hindbrain segmentations within the brain vasculature (top). Moreover, VISTA-Z can perform ISV analysis, and region-based segmentation at the trunk (bottom). \u003cstrong\u003ec\u003c/strong\u003e) Vessel segmentation workflow comprising Meijering-based filtering (1) and initial mask segmentation (2), labelling vessel segments and manual curation (3), and output of the refined mask (4). \u003cstrong\u003ed\u003c/strong\u003e) Measurement of vascular metrics, including skeletonisation (1), distance-based filtered branchpoint analysis (2), vessel diameter (3), and overall network length and vessel density (4). \u003cstrong\u003ee\u003c/strong\u003e) Data filtering and normalisation steps, including MAD-based filtering (1), spatial unit transformation (2) and z-stack normalisation (3). These steps are performed before results are exported to an Excel spreadsheet (4). CLAHE: contrast limited adaptive histogram equalisation, ISV: intersegmental vessel, MAD: median absolute deviation and ROI: region of interest. Scale bars = 100mm.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8402098/v1/7e556bb45e3e876535536417.png"},{"id":100866579,"identity":"ea2e4de0-d645-4dee-8fd3-582af698c4a3","added_by":"auto","created_at":"2026-01-22 08:34:44","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":8557197,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVISTA-Z enables quantitative analysis of the zebrafish vasculature and reveals transgene-specific differences in vessel diameter. a\u003c/strong\u003e) Representative confocal image shows the brain vasculature of zebrafish embryos at 3 dpf. Overlay of \u003cem\u003eTg(kdrl:EGFP)\u003c/em\u003e\u003csup\u003e\u003cem\u003es843\u003c/em\u003e\u003c/sup\u003e;\u003cem\u003eTg(kdrl:Hsa.HRAS-mCherry)\u003c/em\u003e\u003csup\u003e\u003cem\u003es916\u003c/em\u003e\u003c/sup\u003e (left) or individual channels (right). \u003cstrong\u003eb - e\u003c/strong\u003e) Quantification of vascular metrics including normalised network length (\u003cstrong\u003eb\u003c/strong\u003e, ns = 0.1602), vessel density (\u003cstrong\u003ec\u003c/strong\u003e, ns = 0.3173), number of branchpoints (\u003cstrong\u003ed\u003c/strong\u003e, ns = 0.4831), and vessel diameter (\u003cstrong\u003ee\u003c/strong\u003e, p = 0.006). \u003cstrong\u003ef\u003c/strong\u003e) Correlation analysis compares vessel diameters measured in \u003cem\u003eTg(kdrl:EGFP)\u003c/em\u003e\u003csup\u003e\u003cem\u003es843\u003c/em\u003e\u003c/sup\u003e or \u003cem\u003eTg(kdrl:Hsa.HRAS-mCherry)\u003c/em\u003e\u003csup\u003e\u003cem\u003es916\u003c/em\u003e\u003c/sup\u003e. The equation of the regression line, R-squared (R\u003csup\u003e2\u003c/sup\u003e), and standard error (SE) are represented in the graph. \u003cstrong\u003eg\u003c/strong\u003e) Representative confocal image shows the trunk vasculature of zebrafish embryos at 3 dpf. Overlay of \u003cem\u003eTg(kdrl:EGFP)\u003c/em\u003e\u003csup\u003e\u003cem\u003es843\u003c/em\u003e\u003c/sup\u003e;\u003cem\u003eTg(kdrl:Hsa.HRAS-mCherry)\u003c/em\u003e\u003csup\u003e\u003cem\u003es916\u003c/em\u003e\u003c/sup\u003e (top) or individual channels (bottom). \u003cstrong\u003eg’\u003c/strong\u003e) Enlarged views from (\u003cstrong\u003eG\u003c/strong\u003e) highlight additional EGFP-labelled vascular regions, such as the SIVP (white arrowheads). \u003cstrong\u003eh - k\u003c/strong\u003e) Quantification of vascular metrics including normalised network length (\u003cstrong\u003eh\u003c/strong\u003e, p = 0.0003), vessel density (\u003cstrong\u003ei\u003c/strong\u003e, p = 0.0072), number of branchpoints (\u003cstrong\u003ej\u003c/strong\u003e, p = 0.0029), and vessel diameter (\u003cstrong\u003ek\u003c/strong\u003e, p = 0.0007). \u003cstrong\u003el\u003c/strong\u003e) Correlation analysis of vessel diameters as described in (\u003cstrong\u003ef\u003c/strong\u003e). Panels represent n = 12 (\u003cstrong\u003eb - e\u003c/strong\u003e) and n = 13 (\u003cstrong\u003eh - k\u003c/strong\u003e) embryos obtained from two independent breeding pairs. Statistical significance was assessed using paired Student’s t-test for all panels. SIVP: sub-intestinal vein plexus. Scale bars = 100mm.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8402098/v1/9bca72f1e9b9d7cf8be3c929.png"},{"id":100949612,"identity":"20d414ca-bd7f-470c-98b0-0c77c742a900","added_by":"auto","created_at":"2026-01-23 07:04:36","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":2217750,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVISTA-Z reliably detects increases in vascular complexity during development. a\u003c/strong\u003e) Diagram illustrating the progressive increase in vascular complexity in the developing zebrafish brain from 3 to 5 dpf. \u003cstrong\u003eb\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eRepresentative confocal images of the brain vasculature of \u003cem\u003eTg(kdrl:EGFP)\u003c/em\u003e\u003csup\u003e\u003cem\u003es843\u003c/em\u003e\u003c/sup\u003e embryos between 3 and 5 dpf. \u003cstrong\u003ec – f\u003c/strong\u003e) Quantification of vascular metrics including normalised network length (\u003cstrong\u003ec\u003c/strong\u003e, p ≤ 0.0001 (3 – 4 dpf), p ≤ 0.0001 (4 – 5 dpf) and p ≤ 0.0001 (3 – 5 dpf)), vessel density (\u003cstrong\u003ed\u003c/strong\u003e, p ≤ 0.0001 (3 – 4 dpf), p = 0.0007 (4 – 5 dpf) and p ≤ 0.0001 (3 – 5 dpf)), number of branchpoints (\u003cstrong\u003ee\u003c/strong\u003e, p = 0.0188 (3 – 4 dpf), ns = 0.8891 (4 – 5 dpf) and p = 0.0065 (3 – 5 dpf)), and vessel diameter (\u003cstrong\u003ef\u003c/strong\u003e, ns = 0.4797 (3 – 4 dpf), ns = 0.7709 (4 – 5 dpf) and ns = 0.8641 (3 – 5 dpf)). \u003cstrong\u003eg\u003c/strong\u003e) Diagram illustrating the progressive increase in trunk vascular complexity from 3 to 5 dpf. \u003cstrong\u003eh\u003c/strong\u003e) Representative confocal images of the trunk vasculature of \u003cem\u003eTg(kdrl:EGFP)\u003c/em\u003e\u003csup\u003e\u003cem\u003es843\u003c/em\u003e\u003c/sup\u003e zebrafish embryos between 3 and 5 dpf. \u003cstrong\u003ei – l\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eQuantification of vascular metrics including normalised network length (\u003cstrong\u003ei\u003c/strong\u003e, p = 0.0007 (3 – 4 dpf), p ≤ 0.0001 (4 – 5 dpf) and p ≤ 0.0001 (3 – 5 dpf)), vessel density (\u003cstrong\u003ej\u003c/strong\u003e, p ≤ 0.0001 (3 – 4 dpf), p ≤ 0.0001 (4 – 5 dpf) and p ≤ 0.0001 (3 – 5 dpf)), number of branchpoints (\u003cstrong\u003ek\u003c/strong\u003e, p ≤ 0.0001 (3 – 4 dpf), p = 0.0074 (4 – 5 dpf) and p ≤ 0.0001 (3 – 5 dpf)), and vessel diameter (\u003cstrong\u003el\u003c/strong\u003e, ns = 0.2077 (3 – 4 dpf), ns = 0.4750 (4 – 5 dpf) and ns = 0.9025 (3 – 5 dpf)). Panels represent n = 9 (3 dpf) and n = 11 (4 – 5 dpf) embryos for brain metrics and n = 19 (3 dpf), n = 14 (4 dpf), and n = 13 (5 dpf) for trunk metrics. All embryos were obtained from two independent breeding pairs. Statistical significance was assessed using one-way ANOVA with Tukey's multiple comparisons correction. Scale bars = 100mm.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8402098/v1/4e9949872917425f1d3d47ab.png"},{"id":100866585,"identity":"b7fac25c-44fb-47f6-a16b-826b05d7a4e9","added_by":"auto","created_at":"2026-01-22 08:34:44","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":5757832,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVISTA-Z detects disrupted development of midbrain and hindbrain blood vessels in \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003efoxc1a\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e mutant embryos. a\u003c/strong\u003e) Representative confocal images of the brain vasculature of zebrafish embryos at 3 dpf showing wild type (left) and \u003cem\u003efoxc1a\u003c/em\u003e\u003csup\u003e\u003cem\u003esh356\u003c/em\u003e\u003c/sup\u003e mutant (right) phenotypes. \u003cstrong\u003eb - e\u003c/strong\u003e) Quantification of vascular metrics including normalised network length (\u003cstrong\u003eb\u003c/strong\u003e, p ≤ 0.0001), vessel density (\u003cstrong\u003ec\u003c/strong\u003e, p ≤ 0.0001), number of branchpoints (\u003cstrong\u003ed\u003c/strong\u003e, p = 0.0104), and vessel diameter (\u003cstrong\u003ee\u003c/strong\u003e, p ≤ 0.0001). \u003cstrong\u003ef\u003c/strong\u003e) Enlarged views of the brain vasculature from (\u003cstrong\u003ea\u003c/strong\u003e) showing midbrain (i and ii) and hindbrain (iii and iv) regions from wild type (left) and \u003cem\u003efoxc1a\u003c/em\u003e\u003csup\u003e\u003cem\u003esh356\u003c/em\u003e\u003c/sup\u003e mutant (right) embryos. Vessel segmentation masks (top) and confocal images (bottom).\u003cstrong\u003e g - j\u003c/strong\u003e) Quantification of vascular metrics in midbrain and hindbrain segments, including normalised network length (\u003cstrong\u003eg\u003c/strong\u003e, p ≤ 0.0001 (midbrain) and p ≤ 0.0001 (hindbrain)), vessel density (\u003cstrong\u003eh\u003c/strong\u003e, p ≤ 0.0001 (midbrain) and p ≤ 0.0001 (hindbrain)), number of branchpoints (\u003cstrong\u003ei\u003c/strong\u003e, p ≤ 0.0001 (midbrain) and p = 0.0031 (hindbrain)), and vessel diameter (\u003cstrong\u003ej\u003c/strong\u003e, ns = 0.8407 (midbrain) and ns = 0.3618 (hindbrain)). All panels represent n = 22 (wild type) and n = 18 (\u003cem\u003efoxc1a\u003c/em\u003e\u003csup\u003e\u003cem\u003esh356\u003c/em\u003e\u003c/sup\u003e) embryos from two independent breeding pairs. Statistical significance was assessed using unpaired Student’s t-test for all panels, except for panels (\u003cstrong\u003ei\u003c/strong\u003e and \u003cstrong\u003ej\u003c/strong\u003e: hindbrain), which used Mann–Whitney U test. ACeV: anterior cerebral vein, BA: basilar artery, CtAs: central arteries, DLV: dorsal longitudinal vein, MCeV: middle cerebral vein, MSV: mesencephalic vein, PCeV: posterior cerebral vein, and PMCtA and MMCtA: posterior and middle mesencephalic central arteries. Scale bars = 100mm.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-8402098/v1/f4d7c815ea59abb9c7de22d8.png"},{"id":100866582,"identity":"e6ae7092-118d-4319-baee-b0808a1e163c","added_by":"auto","created_at":"2026-01-22 08:34:44","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":3200451,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVISTA-Z detects brain and trunk vascular defects in \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003ekdrl\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e mutant embryos. a\u003c/strong\u003e) Representative confocal images of the brain vasculature of zebrafish embryos at 3 dpf, showing wild type (left) and \u003cem\u003ekdrl\u003c/em\u003e\u003csup\u003e\u003cem\u003eqmc313\u003c/em\u003e\u003c/sup\u003e mutant (right) phenotypes. \u003cstrong\u003eb - e\u003c/strong\u003e) Quantification of brain vascular metrics, including normalised network length (\u003cstrong\u003eb\u003c/strong\u003e, p ≤ 0.0001), vessel density (\u003cstrong\u003ec\u003c/strong\u003e, p ≤ 0.0001), number of branchpoints (\u003cstrong\u003ed\u003c/strong\u003e, p ≤ 0.0001), and vessel diameter (\u003cstrong\u003ee\u003c/strong\u003e, p ≤ 0.0001). \u003cstrong\u003ef\u003c/strong\u003e) Representative confocal images of the trunk vasculature of zebrafish embryos at 3 dpf, showing wild type (top) and \u003cem\u003ekdrl\u003c/em\u003e\u003csup\u003e\u003cem\u003eqmc313\u003c/em\u003e\u003c/sup\u003e mutant (bottom) phenotypes. \u003cstrong\u003eg - j\u003c/strong\u003e) Quantification of trunk vascular metrics, including normalised network length (\u003cstrong\u003eg\u003c/strong\u003e, p ≤ 0.0001), vessel density (\u003cstrong\u003eh\u003c/strong\u003e, p ≤ 0.0001), number of branchpoints (\u003cstrong\u003ei\u003c/strong\u003e, p ≤ 0.0001), and vessel diameter (\u003cstrong\u003ej\u003c/strong\u003e, p = 0.0379). All panels represent n = 17 embryos (\u003cstrong\u003eb - e\u003c/strong\u003e,\u003cstrong\u003e f’iii \u003c/strong\u003eand\u003cstrong\u003e g - j\u003c/strong\u003e), except for n = 24 wild type embryos in the brain vascular analysis (\u003cstrong\u003eb - e\u003c/strong\u003e). Embryos were obtained from two independent breeding pairs. Statistical significance was assessed using unpaired Student’s t-test for all panels. Scale bars = 100mm.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-8402098/v1/d06a07885b0686d233213387.png"},{"id":100949443,"identity":"0846356d-d4d7-4b5d-a09e-2e25f654dd2b","added_by":"auto","created_at":"2026-01-23 07:02:28","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":3900061,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVISTA-Z detects widespread hyper-vascularisation in \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eplxnd1\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e crispants. a\u003c/strong\u003e) Representative confocal images of the brain vasculature of zebrafish embryos at 3 dpf, showing universal guide (ug) injected controls (left) and \u003cem\u003eplxnd1\u003c/em\u003e\u003cstrong\u003e \u003c/strong\u003ecrispant (right) phenotypes. \u003cstrong\u003eb - e\u003c/strong\u003e) Quantification of brain vascular metrics, including normalised network length (\u003cstrong\u003eb\u003c/strong\u003e, p ≤ 0.0001), vessel density (\u003cstrong\u003ec\u003c/strong\u003e, p ≤ 0.0001), number of branchpoints (\u003cstrong\u003ed\u003c/strong\u003e, p = 0.0014), and vessel diameter (\u003cstrong\u003ee\u003c/strong\u003e, ns = 0.098). \u003cstrong\u003ef\u003c/strong\u003e) Representative confocal images of the trunk vasculature of zebrafish embryos at 3 dpf, showing control-injected (top) and \u003cem\u003eplxnd1\u003c/em\u003e\u003cstrong\u003e \u003c/strong\u003ecrispant (bottom) phenotypes. \u003cstrong\u003eg - j\u003c/strong\u003e) Quantification of trunk vascular metrics, including normalised network length (\u003cstrong\u003eg\u003c/strong\u003e, p ≤ 0.0001), vessel density (\u003cstrong\u003eh\u003c/strong\u003e, p ≤ 0.0001), number of branchpoints (\u003cstrong\u003ei\u003c/strong\u003e, p ≤ 0.0001), and vessel diameter (\u003cstrong\u003ej\u003c/strong\u003e, p = 0.0011). Panels represent n = 20 embryos, except for n = 18 (\u003cstrong\u003eb - e\u003c/strong\u003e) and n = 22 (\u003cstrong\u003eg - j\u003c/strong\u003e) control-injected embryos in the brain and trunk vascular analysis, respectively. Embryos were obtained from two independent breeding pairs. Statistical significance was assessed using unpaired Student’s t-test for all panels, except for panel (\u003cstrong\u003eb\u003c/strong\u003e), which used Mann–Whitney U test. Scale bars = 100mm.\u003c/p\u003e","description":"","filename":"Figure6.png","url":"https://assets-eu.researchsquare.com/files/rs-8402098/v1/92d0bcd731ff23ced473b1f5.png"},{"id":106343613,"identity":"f87a40ea-062d-4ba3-9816-e8abd558a882","added_by":"auto","created_at":"2026-04-07 16:07:30","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":27838960,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8402098/v1/0abee974-aff8-463e-a7fe-221651dd5c9f.pdf"},{"id":100949450,"identity":"f378f9e9-507e-4986-819f-22afc6fb72dd","added_by":"auto","created_at":"2026-01-23 07:02:32","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1947966,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8402098/v1/0cf7eb5a521d192dc4454fae.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"VISTA-Z: Vascular Imaging and Segmentation for Topology Analysis in Zebrafish","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCardiovascular disease remains the leading cause of mortality worldwide, with vascular dysfunction playing a central role in its pathogenesis [1]. This burden underscores the importance of vascular imaging in both cardiovascular medicine and preclinical research. Vascular imaging provides structural information, including vessel morphology and lumen diameter, that influences blood flow dynamics (haemodynamics) [2]. Applying vascular imaging in preclinical models enables mechanistic studies of angiogenesis, vascular remodelling, and haemodynamic regulation under physiological and pathological conditions [3]. Such insights are essential to advance our understanding of vascular disorders.\u003c/p\u003e\n\u003cp\u003eZebrafish (\u003cem\u003eDanio rerio\u003c/em\u003e) is a powerful vertebrate model for studying vascular development and cardiovascular disease [4]. Around 70% of human genes have zebrafish orthologs, including 82% of known disease genes, making it highly relevant for modelling human pathophysiology [5]. The optical transparency of zebrafish embryos and the ease of generating vascular-specific fluorescent transgenic lines enable real-time \u003cem\u003ein vivo\u003c/em\u003e imaging of the entire vascular network. Combined with external fertilisation, rapid \u003cem\u003eex-utero\u003c/em\u003e development, and the capability to survive without circulation for several days, the zebrafish allows the study of gene function leading to severe vascular phenotypes that are lethal in mammals [6]. These advantages position zebrafish as a key model to investigate angiogenesis, vessel architecture, and vascular malformations [7].\u003c/p\u003e\n\u003cp\u003eDespite advances in imaging, zebrafish vascular imaging often falls short of its full potential because manual analysis remains slow, subjective and error prone. Automated pipelines overcome these barriers by delivering rapid, reproducible, and unbiased quantification of vascular metrics [8]. Historically, most vascular analysis tools were designed for datasets with high signal-to-noise ratios and uniform contrast, such as the retinal vasculature [9]. Several algorithms developed for zebrafish vascular analysis address challenges posed by heterogeneous fluorescence and low signal-to-noise ratios [8,10,11].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhile initial approaches to zebrafish vascular quantification relied on microangiography and manual measurements of selected vessels, this required extensive user input and offered limited scalability [12]. To overcome these constraints, more recent methodologies have introduced automated pipelines for processing fluorescent imaging data. Several of these new algorithms are available through open-source platforms or incorporate graphical user interfaces to improve accessibility and streamline analysis [8,10,11]. Although these developments represent substantial progress, many remain confined to specific anatomical regions or require considerable manual correction. These limitations underscore the need for fully automated, scalable solutions capable of accurately quantifying vascular structures across diverse imaging datasets.\u003c/p\u003e\n\u003cp\u003eTo overcome the limitations of existing approaches, we developed VISTA-Z, an automated pipeline for zebrafish vasculature quantification. This workflow integrates unsupervised contrast adjustment, thresholding for vessel background separation, and filtering techniques that highlight tubular structures to achieve robust segmentation. In addition, VISTA-Z introduces novel functionalities, including segment labelling for manual artefact removal, skeletonisation with branchpoint refinement, and normalisation of metrics to physical units and z-stack depth. These refinements improve biological interpretability by correcting for image and experimental variability. Using VISTA-Z, we quantified the vascular expansion\u0026nbsp;of wild type embryos from 3 to 5 days post-fertilisation (dpf), encompassing both brain and trunk regions. Furthermore, we demonstrate its versatility in detecting vascular abnormalities, such as vessel loss in \u003cem\u003efoxc1a\u003c/em\u003e and \u003cem\u003ekdrl\u003c/em\u003e mutants, and excessive angiogenesis in \u003cem\u003eplxnd1\u003c/em\u003e crispants. Collectively, VISTA-Z provides a comprehensive, high-throughput framework for vascular phenotyping in embryonic zebrafish models.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eDevelopment of VISTA-Z, an automated pipeline for vascular quantification\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eVISTA-Z was developed to enable automated, unbiased quantification of zebrafish vascular metrics from fluorescent data. It is optimised to handle low signal-to-noise ratios and heterogeneous expression patterns (Fig. 1a). The pipeline begins with automated image pre-processing, including contrast enhancement using CLAHE and vessel masking via Otsu thresholding (Fig. 1b). CLAHE locally improves vessel visibility in low-signal regions without amplifying background noise [13], while Otsu thresholding provides consistent, unbiased masking across samples [8]. Vessel structures are then segmented using Meijering filtering (Fig. 1c), selected for its superior performance in preserving vessel connectivity and enhancing complex vascular features [14]. To address the detection of vessel artefacts, our pipeline assigns a unique identifier to each vessel segment, applies a size-based exclusion and allows manual curation, enabling users to remove specific segments if required (Fig. 1c). Structural metrics are extracted through a two-step skeletonisation and quantification workflow (Fig. 1d). Expanding core functionality found within the VesselMetrics package [11], vessel structures are first converted into single-pixel centreline representations. These skeletons preserve the topology of the vascular network enabling quantification of structural metrics such as vessel length and diameter (Fig. 1d). We next implemented custom modules for refined branchpoint filtering (Fig. 1d), followed by metric extraction, converted measurements to spatial units (microns) and normalised them by z-stack depth (Fig. 1e). These improvements ensure consistency across imaging conditions and enhance the biological interpretability of results. Finally, outliers are removed using MAD filtering (Fig. 1e), reducing the influence of extreme values [15]. VISTA-Z supports confocal images with optional region-of-interest (ROI) selection to accommodate diverse imaging conditions (Fig. 1b’), and code is available via GitHub (https://github.com/fishyvessels/VISTA-Z).\u003c/p\u003e\n\u003cp\u003eValidation of VISTA-Z performance using double transgenic zebrafish embryos\u003c/p\u003e\n\u003cp\u003eThe contrast-to-noise ratio varies between vascular-specific fluorescent transgenes, but remains relatively consistent across developmental stages, anatomical regions, and vessel sizes [8]. The use of distinct transgenes provides complementary vascular information that improves segmentation accuracy [16]. It is therefore necessary to validate image analysis algorithms across different fluorescent transgenes and developmental stages to ensure robustness. To validate our pipeline, we employed \u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u003c/sup\u003e\u003c/em\u003e;\u003cem\u003eTg(kdrl:Hsa.HRAS-mCherry)\u003csup\u003es916\u003c/sup\u003e\u003c/em\u003e double transgenic zebrafish embryos, which fluorescently label the cytoplasm and membrane of endothelial cells, respectively. We performed validation across two anatomically distinct vascular areas, the brain and the trunk, at 3 and 5 dpf to account for both anatomical and developmental changes. At 3 dpf, confocal imaging of the brain vasculature revealed comparable network architecture between both reporters (Fig. 2a). No significant differences in normalised network length (Fig. 2b), vessel density (Fig. 2c), and number of branchpoints (Fig. 2d) were observed between the two transgenes. Interestingly, vessel diameters were significantly lower when measured using the \u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u003c/sup\u003e\u003c/em\u003e transgene compared to \u003cem\u003eTg(kdrl:Hsa.HRAS-mCherry)\u003csup\u003es916\u003c/sup\u003e\u003c/em\u003e (Fig. 2e). Linear regression showed a moderate correlation between diameter measurements (Fig. 2f), suggesting differences between transgenes contributed to observed discrepancies in vessel diameter. A non-uniform cytoplasmic localisation of EGFP could be observed on confocal images (Fig. 2a, S1a-b), unlike the membrane localisation of mCherry [17]. Therefore, to determine whether fluorophore localisation could influence diameter measurements, we examined the cellular localisation of EGFP in a \u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u003c/sup\u003e\u003c/em\u003e;\u003cem\u003eTg(fli1a:nls-mCherry)\u003csup\u003esh550\u003c/sup\u003e\u0026nbsp;\u003c/em\u003edouble transgenic zebrafish line. Co-localisation analysis revealed EGFP localisation within the endothelial cell cytoplasm and nucleus (Fig. S1b-c), as previously shown but not reported [8,18,19]. Collectively, this suggests reduced vessel diameters result from differences cytoplasmic EGFP localisation compared to membrane mCherry localisation. Given that the membrane is the outermost part of the cell and that measurements extending to the membrane would be expected to be greater than those extending to the limit of cytoplasm, this suggests VISTA-Z is sufficiently robust to detect small changes in diameter due to differences in fluorophore localisation in distinct transgenes.\u003c/p\u003e\n\u003cp\u003eWe next employed VISTA-Z to capture the stereotypical trunk vascular architecture at 3 dpf (Fig. 2g). Interestingly, normalised network length (Fig. 2h), vessel density (Fig. 2i) and the number of branchpoints (Fig. 2j) were significantly increased using \u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u003c/sup\u003e\u003c/em\u003e compared to \u003cem\u003eTg(kdrl:Hsa.HRAS-mCherry)\u003csup\u003es916\u003c/sup\u003e\u003c/em\u003e. \u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u003c/sup\u003e\u0026nbsp;\u003c/em\u003elabelled the developing sub-intestinal vein plexus (SIVP) in contrast to \u003cem\u003eTg(kdrl:Hsa.HRAS-mCherry)\u003csup\u003es916\u003c/sup\u003e\u003c/em\u003e, which was absent from this region \u0026nbsp;(Fig. 2g-g’), suggesting labelling of these additional vessels by EGFP likely explains increases in the trunk vascular network observed in this transgenic background. Furthermore, similar to the brain vasculature, vessel diameters were significantly lower when measured using \u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u003c/sup\u003e\u003c/em\u003e (Fig. 2k-l), indicating that diameter measurements are particularly sensitive to the subcellular localisation of fluorescent transgenes. These findings indicate that the selection of transgenes is an important parameter that can influence vascular measurements.\u003c/p\u003e\n\u003cp\u003eTo examine the versatility of VISTA-Z across developmental stages, we analysed embryos at 5 dpf when the zebrafish vasculature has undergone further remodelling and maturation [12]. At 5 dpf, both brain and trunk vascular networks displayed increased complexity compared to 3 dpf (Fig. S2a and S2g).\u0026nbsp;No significant differences were observed in normalised network length (Fig. S2b and S2h), vessel density (Fig. S2c and S2i) and number of branchpoints (Fig. S2d and S2j) at this stage, with either transgene in both brain and trunk regions. However, vessel diameter was reduced in \u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u0026nbsp;\u003c/sup\u003e\u003c/em\u003ecompared to \u003cem\u003eTg(kdrl:Hsa.HRAS-mCherry)\u003csup\u003es916\u003c/sup\u003e\u003c/em\u003e (Fig. S2e-f and S2k-l). Together, these results establish the robustness of our pipeline and identify transgene-specific effects that can influence vessel measurements.\u003c/p\u003e\n\u003cp\u003eWe implemented ROI-based quantification in VISTA-Z to compare different vascular regions within the brain at 3 dpf. We first validated our ROI selection by assessing left–right symmetry in \u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u003c/sup\u003e\u003c/em\u003e;\u003cem\u003eTg(kdrl:Hsa.HRAS-mCherry)\u003csup\u003es916\u003c/sup\u003e\u003c/em\u003e double transgenic zebrafish embryos (Fig. S3a). There were no significant differences between hemispheres in all quantified vessel metrics, including overall network length, vessel density, or the number of branchpoints (Fig. S3b-e), consistent with the high degree of brain vascular symmetry reported at this developmental stage [10]. No significant differences in the vasculature were observed between the midbrain and hindbrain (Fig. S3f-j), with the exception that hindbrain vessels showed fewer branchpoints and smaller diameters than midbrain vessels in \u003cem\u003eTg(kdrl:Hsa.HRAS-mCherry)\u003csup\u003es916\u003c/sup\u003e\u003c/em\u003e embryos (Fig. S3i and S3j). These differences likely reflect intrinsic biological variation in vessel architecture between these regions. The hindbrain contains larger vessels, such as the basilar artery (BA), and primordial hindbrain channels (PHBCs), which reduce branching complexity, and thinner segments, such as the central arteries (CtAs),\u0026nbsp;which may influence diameter measurements [12]. Consistent with our previous results, vessel diameters were reduced when measured using \u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u0026nbsp;\u003c/sup\u003e\u003c/em\u003ecompared to \u003cem\u003eTg(kdrl:Hsa.HRAS-mCherry)\u003csup\u003es916\u003c/sup\u003e\u003c/em\u003e (Fig. S3j). Taken together, these results demonstrate the robustness of VISTA-Z in extracting quantitative vascular data from specific regions of the embryo.\u003c/p\u003e\n\u003cp\u003eQuantification of zebrafish vascular network expansion\u003c/p\u003e\n\u003cp\u003eVascular networks increase in complexity as they expand during development [12]. Thus, we next compared different developmental stages to determine whether our pipeline could identify this increase in complexity. We used VISTA-Z to quantify development of the vascular network within the brain and trunk between 3 and 5 dpf. In the brain (Fig. 3a-b), network length (Fig. 3c) and vessel density (Fig. 3d) increased progressively. Branchpoints increased between 3 and 4 dpf but then plateaued (Fig. 3e), indicating that most brain vessels form by 4 dpf, with subsequent growth driven by elongation and refinement rather than new branching, in agreement with previous studies [20]. Vessel diameter remained unchanged across timepoints (Fig. 3f), suggesting that higher vascular density reflects network expansion rather than vessel widening. Similar trends were observed in the trunk vasculature (Fig. 3g-h). From 3 to 5 dpf, normalised network length (Fig. 3i) and vessel density (Fig. 3j) increased due to ongoing angiogenesis and remodelling. Unlike the developing brain, branchpoints increased at each stage between 3 and 5 dpf (Fig. 3k), with sustained trunk angiogenesis and vascular remodelling. As in the developing brain, vessel diameter within the trunk showed no significant variation (Fig. 3l). These findings confirm VISTA-Z’s ability to detect developmental changes in vascular architecture during development and establish a reference for future studies.\u003c/p\u003e\n\u003cp\u003eVISTA-Z detects gross vascular abnormalities in the brain in \u003cem\u003efoxc1a\u003c/em\u003e mutants\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe next tested the ability of VISTA-Z to characterise vascular defects in mutant zebrafish embryos. Forkhead box C1 (FOXC1) is a transcription factor which regulate vascular development [21], with mutations in humans linked to cerebral small vessel disease and stroke [22]. We employed a \u003cem\u003efoxc1a\u003csup\u003esh356\u0026nbsp;\u003c/sup\u003e\u003c/em\u003emutant with previously reported loss of central arteries within the developing brain [23]. Quantitative analysis of confocal imaging of \u003cem\u003efoxc1a\u003csup\u003esh356\u003c/sup\u003e\u003c/em\u003e;\u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u003c/sup\u003e\u003c/em\u003e zebrafish embryos (Fig. 4a) indicated reduced normalised network length (Fig. 4b) and vessel density at 3 dpf (Fig. 4c), reflecting loss of brain vascular segments (Fig. 4a) [23–26]. Branchpoints were also reduced, though with higher variability (Fig. 4d). Notably, vessel diameter increased in \u003cem\u003efoxc1a\u003csup\u003esh356\u0026nbsp;\u003c/sup\u003e\u003c/em\u003emutants (Fig. 4e), suggesting that brain network loss in this mutant may alter blood flow dynamics, influencing vessel calibre. We used the VISTA-Z ROI tool to segment the brain vasculature into midbrain and hindbrain regions using the middle cerebral vein (MCeV) as a boundary (Fig. 4f).\u0026nbsp;Regional analysis highlighted reduced network length\u0026nbsp;(Fig. 4g), vessel density\u0026nbsp;(Fig. 4h)\u0026nbsp;and branchpoints\u0026nbsp;(Fig. 4i)\u0026nbsp;in both midbrain and hindbrain regions of \u003cem\u003efoxc1a\u003c/em\u003e mutants. While vessel diameter showed no significant differences between regions\u0026nbsp;(Fig. 4j), variability was markedly higher than in the overall brain vasculature analysis\u0026nbsp;(Fig. 4e and 4j). This pattern likely reflects the absence of arterial segments such as CtAs, BA, and posterior and middle mesencephalic central arteries (PMCtA and MMCtA) in \u003cem\u003efoxc1a\u003c/em\u003e mutants, whereas venous structures, including MCeVs, posterior cerebral veins (PCeVs), dorsal longitudinal vein (DLV), mesencephalic veins (MSVs), and anterior cerebral vein (ACeV) appeared morphologically normal (Fig. 4f).\u0026nbsp;These findings indicate a preferential requirement and previously unrealised role for \u003cem\u003efoxc1a\u003c/em\u003e in midbrain and hindbrain arterial development. Overall, VISTA-Z is capable of reliably detecting regional vascular abnormalities.\u003c/p\u003e\n\u003cp\u003eVISTA-Z detects global vascular patterning defects in \u003cem\u003ekdrl\u003c/em\u003e mutants\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ekdrl\u003c/em\u003e encodes a zebrafish orthologue of mammalian VEGFR2/KDR (Fig. S4), a receptor tyrosine kinase essential for endothelial cell proliferation, migration, and angiogenic sprouting [27,28].\u0026nbsp;Quantitative analysis of \u003cem\u003ekdrl\u003csup\u003eqmc313\u003c/sup\u003e\u003c/em\u003e;\u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u003c/sup\u003e\u003c/em\u003e zebrafish embryos (Fig. 5a) showed significant reductions in normalised network length (Fig. 5b), vessel density (Fig. 5c), and branchpoints at 3 dpf (Fig. 5d), consistent with extensive cerebrovascular loss in previously reported \u003cem\u003ekdrl\u003c/em\u003e mutants [28,29]. Vessel diameter was increased (Fig. 5e), possibly reflecting compensatory haemodynamic changes in \u003cem\u003ekdrl\u003c/em\u003e mutants. We segmented the midbrain and hindbrain vasculature in \u003cem\u003ekdrl\u003c/em\u003e mutants and compared this to controls (Fig. S5a). Regional quantification revealed reduced network length (Fig. S5b), vessel density (Fig. S5c), and branchpoints (Fig. S5d) in both brain regions. Vessel diameter was not significantly different but trended toward larger diameter in mutants (Fig. S5e). These findings indicate that both arteries and veins were abnormal in \u003cem\u003ekdrl\u003c/em\u003e mutants, underscoring widespread cerebrovascular disruption. In the trunk (Fig. 5f), \u003cem\u003ekdrl\u003c/em\u003e mutants displayed reduced normalised network length (Fig. 5g), vessel density (Fig. 5h), and branchpoints (Fig. 5i) similar to observations in the brain vasculature. In addition, \u003cem\u003ekdrl\u003c/em\u003e mutants displayed increased trunk vessel diameter (Fig. 5j). To map spatial vascular patterning, we applied the VISTA-Z ROI tool to segment the trunk into anterior (ISVs 1-4), medial (ISVs 5-25), and posterior (ISVs 26-30) regions based on the stereotypical position of ISVs in wild type embryos at 3 dpf (Fig. 3h) [28]. The anterior and medial trunk regions showed the most severe phenotype, with the medial trunk displaying near-complete ISV loss (Fig. S6a and S6b), consistent with previous studies [30,31]. In contrast, posterior ISVs formed and connected relatively normally (Fig. S6b), \u0026nbsp; suggesting reduced dependence on \u003cem\u003ekdrl\u003c/em\u003e signalling during angiogenesis in the posterior trunk. Quantitative analysis indicated a significant reduction in network length (Fig. S6d) and branchpoints (Fig. S6f) in the medial trunk, with a trend towards reduction in the anterior region. Vessel density was markedly reduced in the anterior and medial trunk, consistent with ISV loss, whereas the posterior trunk exhibited a more variable and less pronounced reduction in ISVs (Fig. S6e). Vessel diameter was not significantly different when assessed by brain region (Fig. S6g). Collectively, these results demonstrate that VISTA-Z enables detection of regional vascular changes in anatomically and morphologically distinct territories.\u003c/p\u003e\n\u003cp\u003eVISTA-Z reliably quantifies hyper-vascularisation in \u003cem\u003eplxnd1\u003c/em\u003e crispants\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAfter characterising mutants with reduced vascular networks, we applied VISTA-Z to analyse \u003cem\u003eplxnd1\u003c/em\u003e crispants which display elevated vascular network formation (Fig. 6). \u003cem\u003eplxnd1\u003c/em\u003e mutation causes excessive angiogenic sprouting, aberrant vessel branching, and disorganised vascular patterning (Fig. 6a) [32,33]. We generated mosaic \u003cem\u003eplxnd1\u003c/em\u003e crispants via CRISPR–Cas9 in \u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u003c/sup\u003e\u003c/em\u003e embryos. Quantification showed significant increases in normalised network length (Fig. 6b), vessel density (Fig. 6c), and branchpoints (Fig. 6d), indicating excessive vessel formation. Unlike \u003cem\u003efoxc1a\u003c/em\u003e and \u003cem\u003ekdrl\u003c/em\u003e mutants, vessel diameters in \u003cem\u003eplxnd1\u003c/em\u003e crispants remained unchanged in the developing brain (Fig. 6e). To characterise the spatial distribution of vascular overgrowth, we employed VISTA-Z to segment the brain into midbrain and hindbrain regions using the MCeVs as a boundary (Fig. S7a). Quantitative analysis at 3 dpf indicated significant increases in normalised network length (Fig. S7b), vessel density (Fig. S7c), and branchpoints in both regions (Fig. S7d), while vessel diameter was not significantly different between \u003cem\u003eplxnd1\u003c/em\u003e crispants and controls (Fig. S7e). The brain vascular overgrowth observed in \u003cem\u003eplxnd1\u003c/em\u003e crispants is likely due to hyper-angiogenesis of the CtAs, MMCtA and PMCtA and aberrant anastomosis of the PCeV and ACeV (Fig. S7a). These findings suggest that \u003cem\u003eplxnd1\u003c/em\u003e disruption drives ectopic angiogenesis across the brain vasculature. Throughout the trunk (Fig. 6f), \u003cem\u003eplxnd1\u003c/em\u003e crispants showed significant increases in network length (Fig. 6g), vessel density (Fig. 6h), and branchpoints at 3 dpf (Fig. 6i), consistent with pronounced hyper-vascularisation. Vessel diameter decreased (Fig. 6j), likely reflecting immature and unlumenised vessels. Applying VISTA-Z to trunk segmentation (Fig. S8a), all regions (anterior, medial and posterior) showed increased network length (Fig. S8b) and vessel density (Fig. S8c) at 3 dpf, with branchpoints elevated in the anterior and medial trunk but not in the posterior trunk (Fig. S8d). Vessel diameters were not significantly different throughout the trunk in \u003cem\u003eplxnd1\u003c/em\u003e crispants (Fig. S8e). Notably, ventral vascular beds such as the SIVP expanded into normally avascular territories (Fig. S8f). The SIVP of plxnd1 crispants exhibited increased network length (Fig. S8g), vessel density (Fig. S8h), and branchpoints (Fig. S8i), without significant changes in vessel diameter (Fig. S8j). Collectively, these results indicate VISTA-Z detects both vascular expansion and reduction, underscoring its utility for characterising diverse vascular phenotypes.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eQuantitative analysis of vascular architecture is fundamental for understanding angiogenesis and vascular pathology, yet current zebrafish imaging workflows remain constrained by manual segmentation and limited scalability. Existing automated tools have improved accessibility but often lack robustness across diverse imaging conditions, anatomical regions and developmental stages [10,11].\u0026nbsp;VISTA-Z addresses these gaps by introducing a fully integrated pipeline that combines unsupervised vessel segmentation with segment-level labelling, facilitating optional manual curation, branchpoint refinement and systematic filtering and normalisation of vessel metrics, advancing beyond prior frameworks that rely on extensive manual correction.\u003c/p\u003e\n\u003cp\u003eValidation across double-transgenic lines, multiple developmental stages, and different anatomical regions in distinct gain- and loss-of-function zebrafish mutants confirms the robustness of VISTA-Z. We first analysed the embryonic development of the vasculature in transgenic embryos between 3 and 5 dpf, enabling detection of vascular changes during development. These results demonstrate VISTA-Z’s capacity to identify significant changes in vascular architecture across developmental stages and anatomical regions, underscoring its applicability for pre-clinical zebrafish models.\u003c/p\u003e\n\u003cp\u003eAutomated pipelines have transformed vascular image analysis, yet many existing approaches struggle with segmentation accuracy. This often results from misclassifying non-vascular regions or from poor performance in areas with low signal-to-noise contrast [10]. To overcome these limitations, VISTA-Z employs a novel two-step segmentation strategy. First, an unsupervised Meijering-based filter is applied for automated vessel detection [14]. Next, individual segment labelling enables targeted artefact removal, generating high-quality vessel masks with minimal user intervention. Manual curation continues to play a vital role in vascular analysis, especially with complex embryonic datasets, where fluorescent artefacts or incomplete vessel lumenisation can interfere with segmentation [8]. Our two-step approach offers a robust and scalable solution in the absence of established standards for vessel segmentation and vascular analysis [8,11]. The inclusion of segment-level labelling for artefact removal represents a novel approach to improve segmentation accuracy without sacrificing automation, addressing a persistent challenge in embryonic datasets characterised by low signal-to-noise ratios.\u003c/p\u003e\n\u003cp\u003eA key innovation of VISTA-Z lies in its ability to standardise vascular measurements across heterogeneous datasets, enabling reproducible comparisons between experiments and laboratories. Our branchpoint refinement strategy addresses the overestimation of segment connections that arises when 3D vascular structures are projected onto 2D images using maximum intensity projections [11]. Although full 3D reconstruction is ideal, it is often constrained by low signal and imaging artefacts [10,16]. Therefore, we implemented distance-based filtering to reduce the false detection of branchpoints. Moreover, metric normalisation by z-depth and conversion into spatial units standardise measurements across the diverse imaging modalities used in the field. This ensures that differences reflect biological variation rather than technical inconsistencies. Previous frameworks often lacked these corrections, introducing variability and limiting biological interpretability [11,15]. Our refinements mitigate these limitations, improving cross-study reproducibility and enabling improved vascular comparisons.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eValidating image analysis frameworks across multiple fluorescent reporters and anatomical regions is crucial to ensure their robustness [16,34]. Contrast-to-noise ratios have been shown to vary between zebrafish transgenic lines but remain relatively stable across developmental stages and vessel sizes [8]. Using double transgenic lines provides complementary vascular information, improving segmentation accuracy and mitigating artefacts associated with single-reporter analysis [16]. The use of multiple reporters also addresses common issues, such as variation in promoter activity and fluorophore localisation, which can impact vessel measurements and lead to misinterpretation of results [17,35]. Notably, our data show reporter-specific differences in vessel diameter measurements in both the brain and trunk vasculature across different developmental stages. This underscores the importance of considering reporter cellular localisation in experiments, which can influence vessel metrics. The ability of VISTA-Z to detect significant differences in vessel diameter between membrane or cytoplasmic-localised endothelial transgenic lines highlights its sensitivity and potential utility in identifying subtle vascular changes. Validation across anatomically distinct vascular territories further demonstrates the robustness and utility of VISTA-Z \u0026nbsp;to characterise vascular regions which differ in complexity and morphology [12].\u003c/p\u003e\n\u003cp\u003eDefining how morphometric parameters such as network length, branchpoint density, and vessel diameter change during development is essential to distinguish pathological alterations from normal developmental variation [36]. We observed linear increases in network length and branchpoint density, without notable changes in vessel diameter, indicating that vascular expansion in zebrafish between 3 and 5 dpf primarily reflects elongation and branching rather than lumen widening, consistent with angiogenic remodelling [37].\u003c/p\u003e\n\u003cp\u003eApplying VISTA-Z to study the vasculature of genetic mutants highlights its ability to detect diverse angiogenic perturbations. Our results reinforce the role of \u003cem\u003efoxc1a\u003c/em\u003e in arterial specification and angiogenesis [21,23]. FOXC1 is a transcription factor implicated in vascular development and endothelial cell migration [21,38]. While previous studies, including our own, primarily reported CtA defects in the hindbrain of \u003cem\u003efoxc1a\u003c/em\u003e zebrafish mutant embryos [23,26,39], VISTA-Z analysis reveals that the midbrain vasculature is also compromised, indicating that \u003cem\u003efoxc1a\u003c/em\u003e is required more widely within brain vascular patterning. VISTA-Z analysis also uncovered an increase in vessel diameter in \u003cem\u003efoxc1a\u003c/em\u003e mutants, which we were previously unaware of, likely reflecting compensatory haemodynamic adjustments due to reduced network complexity. Comparable compensation occurs in \u003cem\u003ealk1\u003c/em\u003e zebrafish mutants, in which arteriovenous malformations and vessel dilation occur in response to altered shear stress sensing [40,41]. Reduced vascular resistance and abnormal flow in \u003cem\u003ealk1\u003c/em\u003e mutants trigger adaptive increases in vessel diameter\u0026nbsp;[40,42], suggesting similar mechanisms may function in \u003cem\u003efoxc1a\u003c/em\u003e mutants.\u003c/p\u003e\n\u003cp\u003eKdrl is a membrane receptor essential for angiogenic sprouting, endothelial proliferation, migration, and survival, which mediates VEGF signalling [27,30]. Our \u003cem\u003ekdrl\u003csup\u003eqmc313\u003c/sup\u003e\u003c/em\u003e mutants display severe brain vascular defects, including significant midbrain network disruption, consistent with previously reported \u003cem\u003ekdrl\u003c/em\u003e mutants [27–29]. While \u003cem\u003ekdrl\u003c/em\u003e is dispensable for primary vasculogenesis [27,28,43], it is essential to promote elaboration of complex vascular networks via angiogenesis. We quantified ISV defects along the trunk of \u003cem\u003ekdrl\u003c/em\u003e mutant embryos, revealing a pronounced loss in the medial trunk (ISVs 4–25), while posterior ISVs (25–30) were largely preserved, indicating region-specific requirements for \u003cem\u003ekdrl\u003c/em\u003e during trunk angiogenesis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eValidation of VISTA-Z using \u003cem\u003eplxnd1\u003c/em\u003e crispants demonstrates its utility in characterising hyper-angiogenic phenotypes. Plxnd1, a transmembrane receptor for class 3 Semaphorins, is a crucial negative regulator of angiogenic sprouting by both antagonising VEGF signalling and by providing repulsive cues to migrating endothelial tip cells [32,33]. While previous studies focused on its role in trunk vasculature development [32,33,44], our analysis shows that \u003cem\u003eplxnd1\u003c/em\u003e also limits brain angiogenesis. In the trunk, \u003cem\u003eplxnd1\u003c/em\u003e crispants exhibit pronounced vascular overgrowth, along with reduced vessel diameters, likely due to the presence of immature and unlumenised angiogenic sprouts, which lack functional lumens and only expand in diameter as they mature and experience blood flow [45].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eVISTA-Z provides a robust, automated framework for zebrafish vascular analysis, combining accurate segmentation with standardised metric extraction and regional quantification. By enabling reproducible phenotyping across developmental stages and anatomical regions, this pipeline addresses key limitations of existing tools and offers a scalable solution for high-throughput vascular studies. Collectively, these findings highlight VISTA-Z's ability to resolve spatially distinct vascular phenotypes and to uncover mechanisms that may be overlooked by manual approaches.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cp\u003eZebrafish strains and husbandry\u003c/p\u003e\n\u003cp\u003eZebrafish husbandry and experimental procedures were approved by the University of Nottingham Animal Welfare and Ethical Review Body and conducted under UK Home Office Licence 4626831. All methods were performed in accordance with institutional and national animal welfare guidelines, Directive 2010/63/EU of the European Parliament, and ARRIVE guidelines on the protection of animals used for scientific purposes. The following published lines were employed: \u003cem\u003eTg(kdrl:EGFP)\u003csup\u003es843\u003c/sup\u003e\u003c/em\u003e [46],\u003cem\u003e\u0026nbsp;Tg(kdrl:Hsa.HRAS-mCherry)\u003csup\u003es916\u003c/sup\u003e\u003c/em\u003e [17], \u003cem\u003eTg(fli1a:nls-mCherry)\u003csup\u003esh550\u003c/sup\u003e\u003c/em\u003e [47] and \u003cem\u003efoxc1a\u003csup\u003esh356\u003c/sup\u003e\u003c/em\u003e mutants [23], and \u003cem\u003enacre\u003csup\u003ew2\u003c/sup\u003e\u003c/em\u003e mutants [48] to facilitate confocal imaging within these transgenic and mutant backgrounds. Adult zebrafish were housed following standard husbandry protocols [49]. Embryos were obtained by controlled pair-mating and maintained in E3 buffer (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, and 0.33 mM MgSO\u003csub\u003e4\u003c/sub\u003e). All embryo imaging was performed under anaesthesia using tricaine at 160 mg/mL, and euthanasia was performed by extended incubation in this solution until permanent cessation of circulation.\u003c/p\u003e\n\u003cp\u003eGeneration of a novel \u003cem\u003ekdrl\u003c/em\u003e mutant allele\u003c/p\u003e\n\u003cp\u003eCRISPR Cas9 mutagenesis was performed as described [50]. Briefly, a predesigned Alt-R™ CRISPR RNA (crRNA) (IDT) targeting \u003cem\u003ekdrl\u003c/em\u003e (ENSDARG00000105215) at 5’-GGGACCGTGAAACCTTTTTTTGG-3’ was annealed with trans-activating CRISPR RNA (tracrRNA) (IDT) to form\u0026nbsp;single guide RNA (sgRNA), which was complexed with Engen\u003csup\u003e®\u003c/sup\u003e Spy Cas9 NLS protein (NEB) and injected into one-cell stage embryos [50]. sgRNA efficiency was assessed by high-resolution melt analysis using the following primers: forward: 5’-TGGCAGATGATGTTAAAAGAGGG-3’ reverse: 5’-CACCCTGGTCAAGCATGTAA-3’. G0 CRISPR Cas9-injected adults were incrossed, and progeny were genotyped for \u003cem\u003ekdrl\u003csup\u003eqmc313\u0026nbsp;\u003c/sup\u003e\u003c/em\u003eallele as described [51]. \u003cem\u003ekdrl\u003csup\u003eqmc313\u003c/sup\u003e\u003c/em\u003e is a 2bp deletion\u0026nbsp;within exon 5, which destroys a BslI restriction site and is predicted to prematurely truncate before the conserved transmembrane and intracellular tyrosine kinase domains [27].\u003c/p\u003e\n\u003cp\u003eGeneration of \u003cem\u003eplxnd1\u003c/em\u003e G0 crispants\u003c/p\u003e\n\u003cp\u003eCRISPR Cas9 mutagenesis was carried out as described above. Guide RNAs (crRNAs) were designed using the Alt-R™ CRISPR-Cas9 system (IDT) to target exon 1 of \u003cem\u003eplxnd1\u003c/em\u003e (ENSDARG00000086057). The selected target sequence was 5’-GGTGCTCGCGTTCTCGTGG-3’ [52]. crRNA and tracrRNA were annealed and complexed with Engen\u003csup\u003e®\u003c/sup\u003e Spy Cas9 NLS protein (NEB) before injection into one-cell stage embryos [50]. Crispants were identified based on the previously described vascular phenotype, characterised by excessive sprouting extending beyond intersegmental vessel (ISV)-somite boundaries [32,33,52].\u003c/p\u003e\n\u003cp\u003eConfocal imaging\u003c/p\u003e\n\u003cp\u003eFor live confocal imaging, embryos were anaesthetised in tricaine, embedded in 0.7% low-melting point agarose (Fisher Scientific) in E3 buffer and mounted on 27 mm Nunc™ glass bottom dishes (ThermoFisher). Imaging was performed on a Zeiss LSM880 confocal microscope with Zen Black software (Zeiss) using 10x (NA 0.3) and 20x (NA 0.8) air objectives. All images are 2D maximum intensity projections of 3D z-stacks acquired with a 5 mm step size. Data were acquired using line-sequential scanning mode at an XY resolution of 0.83 mm/pixel, and images were captured as 16-bit data (1024x1024). Image processing and analysis were performed using Python (version 3.10.11); code is available on GitHub: https://github.com/fishyvessels/VISTA-Z\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eImage pre-processing and segmentation\u003c/p\u003e\n\u003cp\u003eMaximum intensity projections were obtained using the CziFile library (version 2019.7.2.1) [53]. For two-colour acquisitions, channels were processed independently before analysis. Images were pre-processed by normalising to an 8-bit format and enhancing contrast using Contrast Limited Adaptive Histogram Equalisation (CLAHE; clip limit = 2.0, tile grid size = 4×4) implemented in OpenCV (version 4.11.0) [54] to improve vessel visibility across heterogeneous backgrounds. Initial binary masks were generated using Otsu thresholding [55]. Segmentation of vascular structures was performed using Meijering filtering (segmentation threshold: 10, sigma range: 3-8 pixels, step size: 1 pixel), followed by morphological post-processing to remove small artefacts (\u0026lt;50 pixels) and fill holes (\u0026lt;200 pixels). Our pipeline increases functionality present in the VesselMetrics Python package [11], by introducing custom modules for segment labelling and user-based quality control. Vessel regions were automatically labelled using connected component analysis and visualised using a pseudo-colour palette. Segmentation results were manually curated through an interactive quality control process where non-vascular artefacts were identified by visual inspection and removed from subsequent analysis. These additions improve segmentation accuracy and downstream analysis.\u003c/p\u003e\n\u003cp\u003eAutomated image-based skeleton, vessel metric analysis and outlier filtering\u003c/p\u003e\n\u003cp\u003eQuantitative analysis of the zebrafish vasculature was performed using a combination of core functions from the VesselMetrics Python package [11] and our novel modules, including refined branchpoint filtering and metric normalisation to improve biological relevance and scalability in zebrafish embryo datasets. Briefly,\u0026nbsp;segmented vessel masks were converted into simplified centreline representations using morphological thinning to reduce vessels to single-pixel-wide paths. Branchpoints were identified as skeleton pixels with more than three connected neighbours in a 3x3 connectivity matrix. To remove detection artefacts from clustered branchpoints, we implemented a distance-based filtering approach. Pairwise Euclidean distances were computed between all detected branchpoints, and only the first point within a minimum-distance threshold (mean vessel length ≈ 30 pixels) was retained. Vessel segments were defined as skeleton regions between branchpoints and endpoints, identified through connected component analysis. Vessel diameters were estimated by calculating the maximum inscribed circle diameter at each skeleton pixel using distance transform methods. Individual vascular segment lengths were calculated as the sum of skeleton pixels within each connected segment. Network-level metrics included vascular length (total skeleton length), vessel density (vessel pixel density computed on tiled regions sized to represent 100 µm\u003csup\u003e2\u003c/sup\u003e windows), and branchpoint density (number of branchpoints per unit area) [11]. All measurements were converted from pixels to physical units (microns) using an empirically determined pixel-to-micron scale factor of 1.2 pixels/µm (using Zeiss LSM880 confocal microscope with a 10x (NA 0.3) air objective). Outlier removal was applied to diameter and vessel length distributions using a median absolute deviation-based filtering (MAD) [56,57], with asymmetric thresholds (lower = 1 × MAD, upper = 3 × MAD). Results were exported as Excel files for downstream data visualisation and statistical analysis.\u003c/p\u003e\n\u003cp\u003eStatistical analysis\u003c/p\u003e\n\u003cp\u003eStatistical analysis was performed using GraphPad Prism (10.6.1). Normality was assessed using the Shapiro-Wilk test. Sample sizes (biological replicates from ≥2 breeding pairs) are indicated in figure legends. Data are shown as box-and-whisker plots (median, interquartile range). Two group comparisons used paired or unpaired Student’s t-test (normally distributed data) and Mann–Whitney U test (non-normally distributed data). For ≥3 groups, one- or two-way ANOVA with Tukey’s test (normal) or Kruskal–Wallis with FDR correction (non-normal) was applied. All values were considered significant with p-values ≤ 0.05, (ns, p \u0026gt; 0.05; * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001; **** p ≤ 0.0001). Samples with poor image quality were excluded from analysis.\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCode is available via GitHub (https://github.com/fishyvessels/VISTA-Z). Other data available upon reasonable request from the corresponding author.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by MRC MR/X008215/1 to RNW.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are grateful for confocal imaging support from the University of Nottingham School of Life Sciences Imaging (SLIM) facility, expert zebrafish husbandry assistance from Charlotte Moore and Harry Gilbert.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contribution\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRNW and IRP conceived the project. IRP, JR and RNW designed experiments. IRP and JR performed experiments. IRP and RNW analysed data. IRP and RNW wrote the manuscript and all authors reviewed the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAdditional information - Declaration of competing interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing of interest.\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAlexander Y, Osto E, Schmidt-Trucks\u0026auml;ss A, Shechter M, Trifunovic D, Duncker DJ, et al. Endothelial function in cardiovascular medicine: a consensus paper of the European Society of Cardiology Working Groups on Atherosclerosis and Vascular Biology, Aorta and Peripheral Vascular Diseases, Coronary Pathophysiology and Microcirculation, and Thrombosis. Cardiovasc Res 2021;117:29\u0026ndash;42. https://doi.org/10.1093/CVR/CVAA085.\u003c/li\u003e\n\u003cli\u003eGhaffari S, Leask RL, Jones EAV. Simultaneous imaging of blood flow dynamics and vascular remodelling during development. Development (Cambridge) 2015;142:4158\u0026ndash;67. https://journals.biologists.com/dev/article/142/23/4158/47121.\u003c/li\u003e\n\u003cli\u003eR\u0026ouml;del CJ, Otten C, Donat S, Louren\u0026ccedil;o M, Fischer D, Kuropka B, et al. Blood Flow Suppresses Vascular Anomalies in a Zebrafish Model of Cerebral Cavernous Malformations. Circ Res 2019;125:e43\u0026ndash;54. https://doi.org/10.1161/CIRCRESAHA.119.315076.\u003c/li\u003e\n\u003cli\u003eWilkinson RN, Van Eeden FJM. The zebrafish as a model of vascular development and disease. Prog Mol Biol Transl Sci 2014;124:93\u0026ndash;122. https://doi.org/10.1016/B978-0-12-386930-2.00005-7.\u003c/li\u003e\n\u003cli\u003eHowe K, Clark MD, Torroja CF, Torrance J, Berthelot C, Muffato M, et al. The zebrafish reference genome sequence and its relationship to the human genome. Nature 2013;496:498\u0026ndash;503. https://doi.org/10.1038/nature12111.\u003c/li\u003e\n\u003cli\u003eGut P, Reischauer S, Stainier DYR, Arnaout R. Little fish, big data: Zebrafish as a model for cardiovascular and metabolic disease. Physiol Rev 2017;97:889\u0026ndash;938. https://doi.org/10.1152/PHYSREV.00038.2016.\u003c/li\u003e\n\u003cli\u003eChico TJA, Ingham PW, Crossman DC. Modeling Cardiovascular Disease in the Zebrafish. Trends Cardiovasc Med 2008;18:150\u0026ndash;155. https://doi.org/10.1016/J.TCM.2008.04.002.\u003c/li\u003e\n\u003cli\u003eKugler E, Plant K, Chico T, Armitage P. Enhancement and Segmentation Workflow for the Developing Zebrafish Vasculature. J Imaging 2019;5:14-27. https://doi.org/10.3390/JIMAGING5010014.\u003c/li\u003e\n\u003cli\u003eDash S, Verma S, Kavita, Khan MS, Wozniak M, Shafi J, et al. A Hybrid Method to Enhance Thick and Thin Vessels for Blood Vessel Segmentation. Diagnostics 2021;11:2017-2036. https://doi.org/10.3390/DIAGNOSTICS11112017.\u003c/li\u003e\n\u003cli\u003e Kugler E, Frost J, Silva V, Plant K, Chhabria K, Chico TJA, et al. Zebrafish vascular quantification: A tool for quantification of three-dimensional zebrafish cerebrovascular architecture by automated image analysis. Development (Cambridge) 2022;149: dev199720-dev199735. https://journals.biologists.com/dev/article/149/3/dev199720/274364.\u003c/li\u003e\n\u003cli\u003e McGarry SD, Adjekukor C, Ahuja S, Greysson-Wong J, Vien I, Rinker KD, et al. Vessel Metrics: A software tool for automated analysis of vascular structure in confocal imaging. Microvasc Res 2024;151:104610\u0026ndash;22. https://doi.org/10.1016/j.mvr.2023.104610.\u003c/li\u003e\n\u003cli\u003e Isogai S, Horiguchi M, Weinstein BM. The Vascular Anatomy of the Developing Zebrafish: An Atlas of Embryonic and Early Larval Development. Dev Biol 2001;230:278\u0026ndash;301. https://doi.org/10.1006/DBIO.2000.9995.\u003c/li\u003e\n\u003cli\u003e da Rocha DA, Barbosa ABL, Guimar\u0026atilde;es DS, Greg\u0026oacute;rio LM, Gomes LHN, da Silva Amorim L, et al. An unsupervised approach to improve contrast and segmentation of blood vessels in retinal images using CLAHE, 2D Gabor wavelet, and morphological operations. Research on Biomedical Engineering 2020;36:67\u0026ndash;75. https://doi.org/10.1007/S42600-019-00032-Z.\u003c/li\u003e\n\u003cli\u003e Lamy J, Merveille O, Kerautret B, Passat N, Vacavant A. Vesselness filters: A survey with benchmarks applied to liver imaging. Int Conf Pattern Recognit 2020:3528\u0026ndash;3535. https://doi.org/10.1109/ICPR48806.2021.9412362.\u003c/li\u003e\n\u003cli\u003e McDowell KP, Berthiaume AA, Tieu T, Hartmann DA, Shih AY. VasoMetrics: unbiased spatiotemporal analysis of microvascular diameter in multi-photon imaging applications. Quant Imaging Med Surg 2021;11:969-983. https://doi.org/10.21037/QIMS-20-920.\u003c/li\u003e\n\u003cli\u003e Daetwyler S, Gü Nther U, Modes CD, Harrington K, Huisken J. Multi-sample SPIM image acquisition, processing and analysis of vascular growth in zebrafish. Development (Cambridge) 2019,6:146\u0026ndash;177. https://doi.org/10.1242/dev.173757.\u003c/li\u003e\n\u003cli\u003e Chi NC, Shaw RM, De Val S, Kang G, Jan LY, Black BL, et al. Foxn4 directly regulates tbx2b expression and atrioventricular canal formation. Genes Dev 2008;22:734\u0026ndash;739. https://doi.org/10.1101/GAD.1629408.\u003c/li\u003e\n\u003cli\u003e Wild R, Klems A, Takamiya M, Hayashi Y, Str\u0026auml;hle U, Ando K, et al. Neuronal sFlt1 and Vegfaa determine venous sprouting and spinal cord vascularization. Nat Commun 2017;8:1\u0026ndash;17. https://doi.org/10.1038/ncomms13991.\u003c/li\u003e\n\u003cli\u003e Schevenels G, Cabochette P, America M, Vandenborne A, De Grande L, Guenther S, et al. A brain-specific angiogenic mechanism enabled by tip\u0026nbsp;cell specialization. Nature 2024;628:1\u0026ndash;9. https://doi.org/10.1038/s41586-024-07283-6.\u003c/li\u003e\n\u003cli\u003e Fujita M, Cha YR, Pham VN, Sakurai A, Roman BL, Silvio Gutkind J, et al. Assembly and patterning of the vascular network of the vertebrate hindbrain. Development 2011;138:1705\u0026ndash;1715. https://doi.org/10.1242/DEV.058776.\u003c/li\u003e\n\u003cli\u003e Kume T. The Cooperative Roles of Foxc1 and Foxc2 in Cardiovascular Development. Adv Exp Med Biol 2009;665:63\u0026ndash;77. https://doi.org/10.1007/978-1-4419-1599-3_5.\u003c/li\u003e\n\u003cli\u003e French CR, Waskiewicz AJ, Lehmann OJ. Mutation of FOXC1 and PITX2 induces cerebral small-vessel disease. J Clin Invest 2014;124:4881-4887. https://doi.org/10.1172/JCI75109.\u003c/li\u003e\n\u003cli\u003e Jiang Z, Evans T, Savage AM, Loose M, Chico TJA, Van Eeden FJM, et al. foxc1a and foxc1b differentially regulate angiogenesis from arteries and veins by modulating Vascular Endothelial Growth Factor signalling. BioRxiv 2018:417931. https://doi.org/10.1101/417931.\u003c/li\u003e\n\u003cli\u003e Skarie JM, Link BA. FoxC1 is essential for vascular basement membrane integrity and hyaloid vessel morphogenesis. Invest Ophthalmol Vis Sci 2009;50:5026-5034. https://doi.org/10.1167/IOVS.09-3447.\u003c/li\u003e\n\u003cli\u003e Whitesell TR, Chrystal PW, Ryu JR, Munsie N, Grosse A, French CR, et al. foxc1 is required for embryonic head vascular smooth muscle differentiation in zebrafish. Dev Biol 2019;453:34\u0026ndash;47. https://doi.org/10.1016/J.YDBIO.2019.06.005.\u003c/li\u003e\n\u003cli\u003e Duan X, Shi Y, Zhao S, Yao L, Sheng J, Liu D. Foxc1a regulates zebrafish vascular integrity and brain vascular development through targeting amotl2a and ctnnb1. Microvasc Res 2022;143:104400. https://doi.org/10.1016/J.MVR.2022.104400.\u003c/li\u003e\n\u003cli\u003e Habeck H, Odenthal J, Walderich B, Maischein HM, Bebber van F, Busch-Nentwich E, et al. Analysis of a Zebrafish VEGF Receptor Mutant Reveals Specific Disruption of Angiogenesis. Current Biology 2002;12:1405\u0026ndash;1412. https://doi.org/10.1016/S0960-9822(02)01044-8.\u003c/li\u003e\n\u003cli\u003e Bussmann J, Wolfe SA, Siekmann AF. Arterial-venous network formation during brain vascularization involves hemodynamic regulation of chemokine signaling. Development (Cambridge) 2011;138:1717\u0026ndash;1726. https://doi.org/10.1242/DEV.059881.\u003c/li\u003e\n\u003cli\u003e Fetsko AR, Sebo DJ, Taylor MR. Brain endothelial cells acquire blood-brain barrier properties in the absence of Vegf-dependent CNS angiogenesis. Dev Biol 2023;494:46\u0026ndash;59. https://doi.org/10.1016/J.YDBIO.2022.11.007.\u003c/li\u003e\n\u003cli\u003e Covassin LD, Villefranc JA, Kacergis MC, Weinstein BM, Lawson ND. Distinct genetic interactions between multiple Vegf receptors are required for development of different blood vessel types in zebrafish. Proc Natl Acad Sci U S A 2006;103:6554-6559. https://doi.org/10.1073/PNAS.0506886103.\u003c/li\u003e\n\u003cli\u003e Covassin LD, Siekmann AF, Kacergis MC, Laver E, Moore JC, Villefranc JA, et al. A genetic screen for vascular mutants in zebrafish reveals dynamic roles for Vegf/Plcg1 signaling during artery development. Dev Biol 2009;329:212-226. https://doi.org/10.1016/J.YDBIO.2009.02.031.\u003c/li\u003e\n\u003cli\u003e Torres-V\u0026aacute;zquez J, Gitler AD, Fraser SD, Berk JD, Pham VN, Fishman MC, et al. Semaphorin-plexin signaling guides patterning of the developing vasculature. Dev Cell 2004;7:117\u0026ndash;123. https://doi.org/10.1016/j.devcel.2004.06.008.\u003c/li\u003e\n\u003cli\u003e Zygmunt T, Gay CM, Blondelle J, Singh MK, Flaherty KMC, Means PC, et al. Semaphorin-PlexinD1 Signaling Limits Angiogenic Potential via the VEGF Decoy Receptor sFlt1. Dev Cell 2011;21:301\u0026ndash;314. https://doi.org/10.1016/j.devcel.2011.06.033.\u003c/li\u003e\n\u003cli\u003e Page DJ, Thuret R, Venkatraman L, Takahashi T, Bentley K, Herbert SP. Positive Feedback Defines the Timing, Magnitude, and Robustness of Angiogenesis. Cell Rep 2019;27:3139-3151. https://doi.org/10.1016/J.CELREP.2019.05.052.\u003c/li\u003e\n\u003cli\u003e Victoria AS, Roy B, Victoria LH. Development of an ImageJ-based method for analysing the developing zebrafish vasculature. Vasc Cell 2017;9:21-40. https://doi.org/10.24238/13221-9-1-172.\u003c/li\u003e\n\u003cli\u003e Martin-Valiente E, Du Y, Goemans C, America M, Zindy E, Adam M, et al. Reversal of cerebrovascular anomalies in a zebrafish model of vein of Galen aneurysm. Nat Cardiovasc Res 2025;4:773\u0026ndash;789. https://doi.org/10.1038/s44161-025-00659-5.\u003c/li\u003e\n\u003cli\u003e Gore A V., Monzo K, Cha YR, Pan W, Weinstein BM. Vascular Development in the Zebrafish. Cold Spring Harb Perspect Med 2012;2:a006684. https://doi.org/10.1101/CSHPERSPECT.A006684.\u003c/li\u003e\n\u003cli\u003e Hayashi H, Kume T. Forkhead transcription factors regulate expression of the chemokine receptor CXCR4 in endothelial cells and CXCL12-induced cell migration. Biochem Biophys Res Commun 2008;367:584\u0026ndash;589. https://doi.org/10.1016/J.BBRC.2007.12.183.\u003c/li\u003e\n\u003cli\u003e Topczewska JM, Topczewski J, Solnica-Krezel L, Hogan BLM. Sequence and expression of zebrafish foxc1a and foxc1b, encoding conserved forkhead/winged helix transcription factors. Mech Dev 2001;100:343\u0026ndash;347. https://doi.org/10.1016/S0925-4773(00)00534-7.\u003c/li\u003e\n\u003cli\u003e Corti P, Young S, Chen CY, Patrick MJ, Rochon ER, Pekkan K, et al. Interaction between alk1 and blood flow in the development of arteriovenous malformations. Development (Cambridge) 2011;138:1573\u0026ndash;1582. https://doi.org/10.1242/DEV.060467.\u003c/li\u003e\n\u003cli\u003e Diwan Z, Kang J, Tsztoo E, Siekmann AF. Alk1/Endoglin signaling restricts vein cell size increases in response to hemodynamic cues. Angiogenesis 2024;28:20-35. https://doi.org/10.1007/S10456-024-09955-3.\u003c/li\u003e\n\u003cli\u003e ochon ER, Menon PG, Roman BL. Alk1 controls arterial endothelial cell migration in lumenized vessels. Development (Cambridge) 2016;143:2593\u0026ndash;2602. https://journals.biologists.com/dev/article/143/14/2593/47581.\u003c/li\u003e\n\u003cli\u003e Bahary N, Goishi K, Stuckenholz C, Weber G, LeBlanc J, Schafer CA, et al. Duplicate VegfA genes and orthologues of the KDR receptor tyrosine kinase family mediate vascular development in the zebrafish. Blood 2007;110:3627\u0026ndash;3636. https://doi.org/10.1182/BLOOD-2006-04-016378.\u003c/li\u003e\n\u003cli\u003e Childs S, Chen JN, Garrity DM, Fishman MC. Patterning of angiogenesis in the zebrafish embryo. Development (Cambridge) 2002;129:973\u0026ndash;982. https://doi.org/10.1242/DEV.129.4.973.\u003c/li\u003e\n\u003cli\u003e Gebala V, Collins R, Geudens I, Phng LK, Gerhardt H. Blood flow drives lumen formation by inverse membrane blebbing during angiogenesis in\u0026nbsp;vivo. Nat Cell Biol 2016;18:443\u0026ndash;450. https://doi.org/10.1038/ncb3320.\u003c/li\u003e\n\u003cli\u003e Beis D, Bartman T, Jin SW, Scott IC, D\u0026rsquo;Amico LA, Ober EA, et al. Genetic and cellular analyses of zebrafish atrioventricular cushion and valve development. Development (Cambridge) 2005;132:4193\u0026ndash;4204. https://doi.org/10.1242/DEV.01970.\u003c/li\u003e\n\u003cli\u003e Chen Y, Savage AM, Jiang Z, Kim HR, Evans PC, Wilkinson RN. Blood flow coordinates collective endothelial cell migration during vascular plexus formation and promotes angiogenic sprout regression via vegfr3/flt4. BioRxiv 2025. https://doi.org/10.1101/2021.07.23.453496.\u003c/li\u003e\n\u003cli\u003e Lister JA, Robertson CP, Lepage T, Johnson SL, Raible DW. nacre encodes a zebrafish microphthalmia-related protein that regulates neural-crest-derived pigment cell fate. Development (Cambridge) 1999;126:3757\u0026ndash;3767. https://doi.org/10.1242/DEV.126.17.3757.\u003c/li\u003e\n\u003cli\u003e Alestr\u0026ouml;m P, D\u0026rsquo;Angelo L, Midtlyng PJ, Schorderet DF, Schulte-Merker S, Sohm F, et al. Zebrafish: Housing and husbandry recommendations. Lab Anim 2020;54:213\u0026ndash;224. https://doi.org/10.1177/0023677219869037.\u003c/li\u003e\n\u003cli\u003e Kroll F, Powell GT, Ghosh M, Gestri G, Antinucci P, Hearn TJ, et al. A simple and effective f0 knockout method for rapid screening of behaviour and other complex phenotypes. Elife 2021;10:1\u0026ndash;34. https://doi.org/10.7554/ELIFE.59683.\u003c/li\u003e\n\u003cli\u003e Wilkinson RN, Elworthy S, Ingham PW, van Eeden FJM. A Method for High-Throughput PCR-Based Genotyping of Larval Zebrafish Tail Biopsies. Biotechniques 2013;55:314\u0026ndash;316. https://doi.org/10.2144/000114116.\u003c/li\u003e\n\u003cli\u003e Davis AE, Castranova D, Weinstein BM. Rapid Generation of Pigment Free, Immobile Zebrafish Embryos and Larvae in Any Genetic Background Using CRISPR-Cas9 dgRNPs. Zebrafish 2021;18:242-250. https://doi.org/10.1089/ZEB.2021.0011.\u003c/li\u003e\n\u003cli\u003e Gohlke C. czifile: Read Carl Zeiss image files (CZI). 2013. https://github.com/cgohlke/czifile.\u003c/li\u003e\n\u003cli\u003e Bradski G. The Opencv Library. Doctor Dobb\u0026rsquo;s Journal of Software Tools, 2000;120;122-125. https://drdobbs.com/ftp/2000/2000_11/opencv.txt.\u003c/li\u003e\n\u003cli\u003e Otsu N. A Threshold Selection Method from Gray-Level Histograms. IEEE Trans Syst Man Cybern 1979;SMC-9:62\u0026ndash;66. https://doi.org/10.1109/TSMC.1979.4310076.\u003c/li\u003e\n\u003cli\u003e Rousseeuw PJ, Croux C. Alternatives to the Median Absolute Deviation. J Am Stat Assoc 1993;88:1272-1283. https://doi.org/10.2307/2291267.\u003c/li\u003e\n\u003cli\u003e Leys C, Ley C, Klein O, Bernard P, Licata L. Detecting outliers: Do not use standard deviation around the mean, use absolute deviation around the median. J Exp Soc Psychol 2013;49:764\u0026ndash;766. https://doi.org/10.1016/J.JESP.2013.03.013\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8402098/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8402098/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"High-resolution vascular imaging in zebrafish embryos offers unparalleled insight into angiogenesis, yet quantitative analysis remains limited by manual workflows and inconsistent segmentation. To overcome this, we developed VISTA-Z, an automated Python-based pipeline designed to standardise vascular quantification across diverse fluorescent datasets. The workflow combines adaptive contrast enhancement, vessel segmentation using Meijering filtering, artefact removal through segment labelling, and skeleton-based topology analysis with refined branchpoint detection. Metrics are normalised to physical units and imaging depth, enabling reproducible comparisons across experiments. We validated VISTA-Z using multiple endothelial transgenic lines and developmental stages. The pipeline detected subtle and severe phenotypes, including brain vessel loss in foxc1a mutants, brain and trunk vessel loss in kdrl mutants and widespread hyper-angiogenesis in plxnd1 crispants. VISTA-Z is an open-source, scalable platform for reproducible high-throughput quantification of zebrafish vascular architecture, providing a standardised framework for developmental research and preclinical screening.","manuscriptTitle":"VISTA-Z: Vascular Imaging and Segmentation for Topology Analysis in Zebrafish","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-22 08:34:39","doi":"10.21203/rs.3.rs-8402098/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-03T07:57:39+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-01T23:59:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"118617110684923281468651963538221516255","date":"2026-01-22T17:20:13+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-20T18:10:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"246390905141544966780028159005653288069","date":"2026-01-20T17:15:46+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-20T16:56:18+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-20T16:46:12+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-01-13T13:06:42+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-08T09:45:00+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-01-08T09:20:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e50fbacd-7ff3-4ee9-8f9b-e0e8f50acddf","owner":[],"postedDate":"January 22nd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":61530866,"name":"Biological sciences/Biological techniques"},{"id":61530867,"name":"Biological sciences/Biotechnology"},{"id":61530868,"name":"Biological sciences/Computational biology and bioinformatics"}],"tags":[],"updatedAt":"2026-04-07T16:03:33+00:00","versionOfRecord":{"articleIdentity":"rs-8402098","link":"https://doi.org/10.1038/s41598-026-43301-5","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2026-04-01 16:00:00","publishedOnDateReadable":"April 1st, 2026"},"versionCreatedAt":"2026-01-22 08:34:39","video":"","vorDoi":"10.1038/s41598-026-43301-5","vorDoiUrl":"https://doi.org/10.1038/s41598-026-43301-5","workflowStages":[]},"version":"v1","identity":"rs-8402098","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8402098","identity":"rs-8402098","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}