Three-dimensional visualization of the vascular bundle in a branched bamboo node

Front Plant Sci. 2023 Oct 25:14:1256772. doi: 10.3389/fpls.2023.1256772. eCollection 2023.

Abstract

Bamboo is a natural vascular bundle (VB) reinforced composite material used in more than 10 fields such as construction and furniture. The nodes in bamboo are crucial to its mechanical properties, but understanding of its performance is limited by lack of knowledge about the three-dimensional (3D) structure of the node. This work aimed to non-destructively identify the multi-dimensional characteristics of the VB in a bamboo branched node (BN) using X-ray microtomography (µCT). The VB was segmented from the BN using deep learning combined with the Watershed algorithm. The 3D model reconstruction and characterization of the VB were also conducted. It was found that the structure of VBs showed significant changes along the height of the BN. The VBs formed a complex 3D structure, VBs of the culm are connected with those of the branch, and the connectivity of the conducting tissue and fibers was 88.91% and 99.95%, respectively. The conducting tissue and the fibers had similar shapes but varying thicknesses, which enabled VBs to perform both water transport and mechanical support functions. The volumes fraction of parenchyma, fibers, and conducting tissue in the BN were 61.3%, 35.3%, and 3.4%, respectively, but the tissue proportion of the different heights of the BN varied from each other. The nodal ridge was a mechanical weak point of the BN, with a maximum fibers proportion of 43.8%. This study contributes to understanding the relationship of VBs between the branch and the culm. It provides a structural perspective for understanding the mechanical properties of BN and a theoretical basis for optimizing bamboo utilization efficiency.

Keywords: X-ray microtomography; bamboo; branched node; deep-learning algorithms; three-dimensional reconstruction; vascular bundle.

Grants and funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. National Key Research & Development Program of China (2022YFD2200901); National Natural Science Foundation of China (31670565).