Research Article: Quanfima: An open source Python package for automated fiber analysis of biomaterials

Date Published: April 11, 2019

Publisher: Public Library of Science

Author(s): Roman Shkarin, Andrei Shkarin, Svetlana Shkarina, Angelica Cecilia, Roman A. Surmenev, Maria A. Surmeneva, Venera Weinhardt, Tilo Baumbach, Ralf Mikut, Nuno Araujo.


Hybrid 3D scaffolds composed of different biomaterials with fibrous structure or enriched with different inclusions (i.e., nano- and microparticles) have already demonstrated their positive effect on cell integration and regeneration. The analysis of fibers in hybrid biomaterials, especially in a 3D space is often difficult due to their various diameters (from micro to nanoscale) and compositions. Though biomaterials processing workflows are implemented, there are no software tools for fiber analysis that can be easily integrated into such workflows. Due to the demand for reproducible science with Jupyter notebooks and the broad use of the Python programming language, we have developed the new Python package quanfima offering a complete analysis of hybrid biomaterials, that include the determination of fiber orientation, fiber and/or particle diameter and porosity. Here, we evaluate the provided tensor-based approach on a range of generated datasets under various noise conditions. Also, we show its application to the X-ray tomography datasets of polycaprolactone fibrous scaffolds pure and containing silicate-substituted hydroxyapatite microparticles, hydrogels enriched with bioglass contained strontium and alpha-tricalcium phosphate microparticles for bone tissue engineering and porous cryogel 3D scaffold for pancreatic cell culturing. The results obtained with the help of the developed package demonstrated high accuracy and performance of orientation, fibers and microparticles diameter and porosity analysis.

Partial Text

Biomaterials are often designed to mimic chemical and physical properties, for instance shape, of biological systems [1–3]. The development of special fibrous and porous three-dimensional (3D) structures, so-called scaffolds, has gained popularity in the field of tissue engineering (TE) [4–6]. Such structures can replace and treat damaged body tissues[5]. The detailed analysis of the fibrous structure is essential to reveal dependencies between biomaterial properties and its performance in a tissue. For instance, controlling the fiber orientation in the scaffolds fabrication process allows for advanced customized solutions that promote faster and higher-quality treatment in many fields of TE. For bones, TE requires scaffolds with both, randomly oriented and aligned structures to mimic a native extracellular matrix (ECM) and to ensure appropriate mechanical properties [7]. In contrast, scaffolds designed for nerve and blood vessels are aimed to recreate the natural architecture of tissues with aligned fibers as closely as possible [8,9]. Such property as the fiber diameter influences cell adhesion and growth kinetics [10–12]. Moreover, some scaffolds consist of bioactive particles with different size, that influence the porosity and efficiency properties of the matrix. The porosity of biomaterials is linked to the success of tissue ingrowth [13–15]. The development of biomaterials with desired properties requires 3D characterization of their structure with a precise, ideally automatic, computational analysis.

The most frequently cited software tools for analysis of fibrous materials are compiled in Table 1, where functionality, easiness of software extensibility, and integration with other software by users without long programming experience are presented. FiberScout was developed for analysis of industrial CT datasets of fiber-reinforced materials [63]. It provides a wide range of features, which includes displaying a distribution of fiber lengths and orientation to clustering fibers with equivalent properties and selection of individual fibers by certain criteria. This tool is represented by a module for the open_iA platform, which is written in C++ and uses the ITK library [64]. Among the plug-ins for ImageJ, DiameterJ [41] was developed with a focus on the analysis of SEM micrograph images, providing information about orientation and diameter distributions. The OrientationJ [43] plug-in is aimed initially at the analysis of microscopy images and provides information about orientation, coherency, and energy within a specified ROI. Additionally, it can combine extracted data into a single image to enhance the visual perception. The FibrilTool [42] plug-in has been recently proposed for the study of fibrillar structures of microscopy images, extracting the information of average orientation along with an anisotropy value for a given region of interest (ROI). All these plug-ins are written in Java and run in a Java Virtual Machine, and cannot be easily integrated into the Python environment.

The quanfima package is based on several widespread third-party packages developed for data analysis, such as NumPy, SciPy, scikit-image, VIGRA, Pandas, and Statsmodels. We combined their capabilities to create a complete pipeline for characterization of biomaterials. The functionality of quanfima is separated into four modules (Fig 2).

To test the performance of the developed package, we use a set of generated datasets as well as several types of biomaterials. The cases selected here represent various types of widespread composite 3D scaffolds made from different biomaterials such as polymers, calcium phosphate ceramics, and bioglass at distinct levels of characterization, such as after production and in vitro analysis. While all of the data was acquired with micro-CT, the package can be equally applied to any imaging technique.

We have presented the quanfima package for a complete, comprehensive 2D and 3D analysis of fibrous biomaterials. It provides capabilities for morphological analysis of fibrous structures and offers statistical analysis and visualization without bindings to a specific visualization system. The package can be easily applied to other biomaterials for characterization of wall thicknesses, inclusions, and porosity. The power of such analyses was demonstrated on various biomaterials, such as pure and composite PCL scaffolds, hydrogels and cryogels. In all cases, it provides quantitative 3D analysis and effortless visualization of the results. These examples are supplemented with the Jupyter notebook, making further use of the quanfima package possible as plug and play analysis for any datasets.