Evaluating LEAF GUI versus ImageJ for leaf vein density measurement

Evaluating LEAF GUI versus ImageJ for leaf vein density measurement | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 1 February 2025 V1 Latest version Share on Evaluating LEAF GUI versus ImageJ for leaf vein density measurement Authors : Xuchen Guo 0000-0003-0242-2170 , Qingyue Miao , Yuanmiao Chen , and Jianhui Xue [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.173841676.68117087/v1 405 views 190 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Leaf vein density (LVD) is a critical trait linked to leaf hydraulic efficiency, commonly quantified using an automated tool—LEAF GUI, which analyzes vein structures via pixel-based algorithms. However, concerns persist about its accuracy for species with complex hierarchical venation networks. To evaluate the reliability of LEAF GUI, we compared its LVD measurements against those from ImageJ—a manual tracing platform renowned for its precision—using leaf specimens from nine Magnoliaceae species. Paired t-tests revealed no statistically significant differences between the two methods (P = 0.534), demonstrating comparable accuracy under standardized conditions. However, LEAF GUI’s reliability is constrained by its reliance on laborious threshold calibration and stringent image quality standards. For studies prioritizing precision, especially in taxa with heterogeneous or low-resolution samples, we recommend ImageJ as the standard approach. Its manual tracing protocol achieves consistency in resolving complex vein networks, balancing analytical rigor with adaptability to diverse sample conditions not-yet-known not-yet-known not-yet-known unknown Introduction Leaf vein density (LVD) is a pivotal trait governing hydraulic efficiency in plants, critically shaping water transport pathways, gas exchange dynamics, and photosynthetic performance (Chapin et al., 2002; Blonder et al., 2011; Sack et al., 2012; Jordan et al., 2013). Higher LVD correlates with shorter hydraulic resistance and enhanced rates of transpiration and nutrient uptake, positioning it as a key metric for understanding leaf adaptation to environmental stressors like drought and the evolutionary drivers of venation architecture (Sack and Frole, 2006; Ye et al., 2021; Peng et al., 2022). The automated tool LEAF GUI has gained prominence for quantifying LVD by analyzing venation patterns via pixel-based thresholding algorithms (Price et al., 2011; Price, 2012). However, its accuracy remains contentious for hierarchically organized venation systems (1° primary, 2° secondary, and 3° tertiary veins) that exhibit structural complexity and diversity (Hickey, 1973; Ellis, 2009). While primary and secondary veins are readily distinguishable, higher-order veins (4° and beyond) are often obscured or misrepresented at lower magnifications, leading to measurement inaccuracies (McKown et al., 2010; Sack and Scoffoni, 2013). Price et al. (2014) attribute discrepancies in LEAF GUI-derived LVD values to image magnification effects, microscale lattice geometry, and the fractal nature of vein networks. Conversely, Sack et al. (2014) argue that these biases reflect methodological limitations rather than inherent properties of LVD, advocating for manual approaches like those implemented in ImageJ to mitigate such errors. We evaluate the strengths and limitations of LEAF GUI and ImageJ for quantifying LVD across nine Magnoliaceae species. These species were selected for their intricate and heterogeneous venation patterns(Shi et al., 2022). ImageJ enables meticulous manual tracing of individual veins (including 3° and 4° orders) through localized sampling (Scoffoni & Sack, 2011). By comparing these methodologies, we aim to resolve critical trade-offs between automation efficiency and precision in capturing hierarchical venation traits, ultimately guiding best practices for LVD quantification in complex plant systems. Materials and Methods Leaf sample Nine species of the Magnoliaceae, all growing in Nanjing, Jiangsu Province, China, were selected for this study. We sampled these species in two adjacent sites: Nanjing Zhongshan Botanical Garden (NZB, 32°05′12′′N, 118°83′47′′E) and Nanjing Forestry University (NFU, 32°07′67′′N, 118°81′36′′E). These two site is adjacent, and the geodesic distance between two sampling sites is 3.4 km. For more information see Table 1. Leaf vein visualization A total of 27 leaves (three per species) were placed in a nylon mesh bag and boiled in a 5–10% (w/v) NaOH solution for 30 minutes using an open stainless-steel pot (ST24P1, Supor; diameter: 28 cm, capacity: 1.2 L). Following boiling, the leaves were rinsed under running water, and epidermal and mesophyll tissues were gently removed using a soft brush. The remaining leaf skeletons were stained with a 0.5% (w/v) aqueous safranin solution. Vein specimens were scanned at 600 dpi using an Epson V550 scanner (Epson Indonesia) on air-dried skeletons. Acquisition of leaf vein data and leaf area The software LEAF GUI developed using the MATLAB platform (Price et al., 2011) was used to obtain leaf vein traits and leaf area. The original RGB images were transformed into black and white binary images, and the adaptive threshold and global threshold were set in LEAF GUI to obtain clear leaf vein images. The thresholds were determined empirically for each leaf image. The ImageJ software version Fiji (National Institutes of Health) was also used to measure leaf vein traits according to the protocol established by (Scoffoni and Sack, 2011). Data analysis A paired t -test (α = 0.05) was performed to evaluate differences in leaf vein density (LVD) measurements obtained via ImageJ and LEAF GUI. We also explored the effect of leaf area on the deviation of LVD values (ImageJ vs. LEAF GUI) by applying quadratic regression. All statistical analyses, including hypothesis testing, data modeling, and graphical visualization, were conducted using R software v4.1.2 (R Core Team, 2022). not-yet-known not-yet-known not-yet-known unknown Results Comparisons of LVD values derived from LEAF GUI and ImageJ showed no statistically significant differences (t = -0.631, P = 0.534; Figure 3). Moreover, leaf size variations do not affect the deviation (ImageJ vs. LEAF GUI) in measuring LVD, with P -values for both the linear coefficient (P 1 = 0.382) and quadratic coefficients (P 2 = 0.68) significantly above 0.05 (Figure 3B). Discussion Our analysis indicates strong consistence between the two methodologies for quantifying leaf vein traits under the tested conditions, corroborates the reliability of LEAF GUI-derived data for quantifying leaf venation. However, LEAF GUI heavily dependence on user-defined threshold inputs introduces unavoidable subjectivity, different threshold value can produce significant different LVD (Figure 3). In such cases, the time efficiency advantage of LEAF GUI diminishes compared to manual ImageJ workflows, as meticulous threshold calibration becomes essential to mitigate errors—a process that offset the tool’s intended efficiency gains (Sack et al., 2014). A critical yet frequently underestimated variable in these analyses is staining uniformity. Threshold-based segmentation methods employed by LEAF GUI, rely on consistent staining intensity to distinguish veins from background tissue. Variations in staining quality may cause faint or discontinuous vein signals during binarization, leading to software detection failures (Figure 3). Such artifacts directly compromise the accuracy of LVD, underscoring the importance of standardized sample preparation protocols to minimize technical variability. Overall, LEAF GUI relies on manual and meticulous threshold value selection, as well as specific resolution levels and high-quality vein staining techniques. These requirements can limit both the precision and processing efficiency of LVD measurements. Although large-scale data collection may mitigate the limitations, we recommend using ImageJ for LVD assessments due to its adaptability and ability to ensure data uniformity. Author contributions XG: Formal analysis (Lead), Funding acquisition (Supporting), Investigation (Supporting), Writing - original draft (Lead). QM: Investigation (Lead), Writing - original draft (Supporting), Writing - review & editing (Supporting). YC: Writing - original draft (Supporting), Writing - review & editing (Supporting). JX: Funding acquisition (Lead), Writing - review & editing (Lead) Funding XG was supported by the Postgraduate Research &Practice Innovation Program of Jiangsu Province (KYCX23_1122). JX was supported by the Evaluation on the biodiversity and carbon sequestration of typical wetland ecosystems in Jiangsu Province (No: LYKJ(2022)02). not-yet-known not-yet-known not-yet-known unknown Acknowledgments The authors thank Peijian Shi for his help during the preparation of this work. not-yet-known not-yet-known not-yet-known unknown Conflict of Interest not-yet-known not-yet-known not-yet-known unknown The authors declare no competing financial interests. Data Availability statement: The raw leaf vein density measurement data are included in Supplementary Table S1. References Blonder, B., C. Violle, L. P. Bentley, and B. J. Enquist. 2011. Venation networks and the origin of the leaf economics spectrum. Ecology Letters 14: 91–100.Chapin, F. S., P. A. Matson, and H. A. Mooney. 2002. Principles of terrestrial ecosystem ecology. Springer, New York.Ellis, B. ed. . 2009. Manual of leaf architecture. Cornell University Press, Ithaca.Hickey, L. J. 1973. Classification of the architecture of Dicotyledonous leaves. American Journal of Botany 60: 17–33.Jordan, G. J., T. J. Brodribb, C. J. Blackman, and P. H. Weston. 2013. Climate drives vein anatomy in Proteaceae. American Journal of Botany 100: 1483–1493.McKown, A. D., H. Cochard, and L. Sack. 2010. Decoding Leaf Hydraulics with a Spatially Explicit Model: Principles of Venation Architecture and Implications for Its Evolution. The American Naturalist 175: 447–460.Peng, G., Y. Xiong, M. Yin, X. Wang, W. Zhou, Z. Cheng, Y.-J. Zhang, and D. Yang. 2022. Leaf Venation Architecture in Relation to Leaf Size Across Leaf Habits and Vein Types in Subtropical Woody Plants. Frontiers in Plant Science 13: 873036.Price, C. A. 2012. LEAF GUI: Analyzing the Geometry of Veins and Areoles Using Image Segmentation Algorithms. In J. Normanly [ed.], High-Throughput Phenotyping in Plants, Methods in Molecular Biology, 41–49. Humana Press, Totowa, NJ.Price, C. A., P. R. T. Munro, and J. S. Weitz. 2014. Estimates of Leaf Vein Density Are Scale Dependent. PLANT PHYSIOLOGY 164: 173–180.Price, C. A., O. Symonova, Y. Mileyko, T. Hilley, and J. S. Weitz. 2011. Leaf Extraction and Analysis Framework Graphical User Interface: Segmenting and Analyzing the Structure of Leaf Veins and Areoles. Plant Physiology 155: 236–245.R Core Team. 2022. R: A language and environment for statistical computing.Sack, L., M. Caringella, C. Scoffoni, C. Mason, M. Rawls, L. Markesteijn, and L. Poorter. 2014. Leaf Vein Length per Unit Area Is Not Intrinsically Dependent on Image Magnification: Avoiding Measurement Artifacts for Accuracy and Precision. Plant Physiology 166: 829–838.Sack, L., and K. Frole. 2006. Leaf structural diversity is related to hydraulic capacity in tropical rain forest trees. Ecology 87: 483–491.Sack, L., and C. Scoffoni. 2013. Leaf venation: structure, function, development, evolution, ecology and applications in the past, present and future. New Phytologist 198: 983–1000.Sack, L., C. Scoffoni, A. D. McKown, K. Frole, M. Rawls, J. C. Havran, H. Tran, and T. Tran. 2012. Developmentally based scaling of leaf venation architecture explains global ecological patterns. Nature Communications 3: 837.Scoffoni, C., and L. Sack. 2011. Quantifying leaf vein traits.Shi, P., Q. Miao, Ü. Niinemets, M. Liu, Y. Li, K. Yu, and K. J. Niklas. 2022. Scaling relationships of leaf vein and areole traits versus leaf size for nine Magnoliaceae species differing in venation density. American Journal of Botany 109: 899–909.Ye, M., M. Wu, H. Zhang, Z. Zhang, and Z. Zhang. 2021. High Leaf Vein Density Promotes Leaf Gas Exchange by Enhancing Leaf Hydraulic Conductance in Oryza sativa L. Plants. Frontiers in Plant Science 12: 693815. Tables and Figures Table 1 Leaf sampling information of the nine Magnoliaceae species. NBG: Nanjing Botanical Garden, Chinese Academy of Sciences; NFU: Nanjing Forestry University Xinzhuang Campus. not-yet-known not-yet-known not-yet-known unknown 1 Magnolia amoena Cheng Deciduous NBG 32°3′27′′N, 118°49′56′′E 31 Jul. 2020 2 Magnolia denudata Desr. Deciduous NFU 32°4′43′′N, 118°48′33′′E 13 Sep. 2019 3 Magnolia soulangeana Soul.‐Bod. Deciduous NBG 32°3′29′′N, 118°49′55′′E 30 Jul. 2019 4 Magnolia tomentosa Thunb. Deciduous NBG 32°3′28′′N, 118°49′55′′E 30 Jul. 2019 5 Michelia cavaleriei var. platypetala (Hand.‐Mazz.) N.H. Xia Evergreen NFU 32°4′48′′N, 118°48′30′′E 26 Aug. 2020 6 Michelia chapensis Dandy Evergreen NBG 32°3′28′′N, 118°49′55′′E 25 Jul. 2020 7 Michelia compressa (Maxim.) Sarg. Gard. et For. Evergreen NBG 32°3′28′′N, 118°49′55′′E 30 Jul. 2019 8 Michelia figo (Lour.) Spreng. Evergreen NFU 32°4′46′′N, 118°48′28′′E 29 Jul. 2020 9 Michelia maudiae Dunn Evergreen NFU 32°4′45′′N, 118°48′25′′E 31 Jul. 2020 not-yet-known not-yet-known not-yet-known unknown Figure 1 Examples of chemically treated safranin‐stained leaf vein images for the nine Magnoliaceae species. Figure 2 Paired t-test comparison of the LVD measured by ImageJ and LEAF GUI (A), and a quadratic regression analysis of the deviation between the ImageJ and LEAF GUI measurements across different leaf areas (B). Note: P 1 represents the significant parameter of the linear term, and P 2 is the significant parameter of the quadratic term. Figure 3 The binary modification of leaf vein images using LEAF GUI threshold segments. Information & Authors Information Version history V1 Version 1 01 February 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords comparative method development none of the above plants statistical terrestrial Authors Affiliations Xuchen Guo 0000-0003-0242-2170 Nanjing Forestry University Collaborative Innovation Center of Sustainable Forestry in Southern China of Jiangsu Province View all articles by this author Qingyue Miao Nanjing Forestry University Collaborative Innovation Center of Sustainable Forestry in Southern China of Jiangsu Province View all articles by this author Yuanmiao Chen Nanjing Forestry University Collaborative Innovation Center of Sustainable Forestry in Southern China of Jiangsu Province View all articles by this author Jianhui Xue [email protected] Nanjing Forestry University Collaborative Innovation Center of Sustainable Forestry in Southern China of Jiangsu Province View all articles by this author Metrics & Citations Metrics Article Usage 405 views 190 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Xuchen Guo, Qingyue Miao, Yuanmiao Chen, et al. Evaluating LEAF GUI versus ImageJ for leaf vein density measurement. Authorea . 01 February 2025. DOI: https://doi.org/10.22541/au.173841676.68117087/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.173841676.68117087/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9ff5ed0b1d5241e2',t:'MTc3OTM5MTA5NQ=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();