A computational method for three-dimensional reconstruction of the microarchitecture of myometrial smooth muscle from histological sections

Background
The fibrous structure of the myometrium has previously been characterised at high resolutions in small tissue samples (< 100 mm3) and at low resolutions (∼500 μm per voxel edge) in whole-organ reconstructions. However, no high-resolution visualisation of the myometrium at the organ level has previously been attained.
Methods and results
We have developed a technique to reconstruct the whole myometrium from serial histological slides, at a resolution of approximately 50 μm per voxel edge. Reconstructions of samples taken from human and rat uteri are presented here, along with histological verification of the reconstructions and detailed investigation of the fibrous structure of these uteri, using a range of tools specifically developed for this analysis. These reconstruction techniques enable the high-resolution rendering of global structure previously observed at lower resolution. Moreover, structures observed previously in small portions of the myometrium can be observed in the context of the whole organ. The reconstructions are in direct correspondence with the original histological slides, which allows the inspection of the anatomical context of any features identified in the three-dimensional reconstructions.
Conclusions and significance
The methods presented here have been used to generate a faithful representation of myometrial smooth muscle at a resolution of ∼50 μm per voxel edge. Characterisation of the smooth muscle structure of the myometrium by means of this technique revealed a detailed view of previously identified global structures in addition to a global view of the microarchitecture. A suite of visualisation tools allows researchers to interrogate the histological microarchitecture. These methods will be applicable to other smooth muscle tissues to analyse fibrous microarchitecture.

研究背景
此前，子宫肌层（myometrium）的纤维结构已在小型组织样本（<100 mm³）中完成高分辨率表征，并在全器官重建中以低分辨率（每个体素（voxel）边长约500 μm）开展特征描述。然而，此前尚未实现器官层面的子宫肌层高分辨率可视化。
方法与结果
本研究开发了一种基于连续组织切片重建全子宫肌层的技术，其体素分辨率可达约50 μm/边。本文展示了取自人类与大鼠子宫的样本重建结果，同时结合为此分析专门开发的一系列工具，对重建结果进行组织学验证，并深入探究这些子宫的纤维结构。该重建技术可实现此前仅能以低分辨率观测到的整体结构的高分辨率渲染；此外，此前仅在子宫肌层小区域中观测到的结构，可在全器官的整体背景下被观测到。重建结果与原始组织切片完全对应，可用于检视三维重建中所识别出的任意特征的解剖学背景。
结论与意义
本文提出的方法可生成每个体素边长约50 μm分辨率下的子宫肌层平滑肌精准复现。借助该技术对子宫肌层平滑肌结构进行表征，不仅可获取微结构的全局视图，还能清晰呈现此前已被识别的全局结构细节。一套配套可视化工具可支持研究人员对组织学微结构进行深入分析。上述方法可推广应用于其他平滑肌组织的纤维微结构分析。