PROJECT SUMMARY Dense connective tissue is composed of an abundant collagen network that is maintained and repaired by fibroblasts. The parent R15 proposal is investigating whether fibroblast-mediated collagen remodeling is governed at the tissue-scale by distortion strain energy. The effect of targeted magnitudes of distortion energy on collagen remodeling will be measured by quantifying alterations in the composition, organization, and mechanical behavior of the scaffolds. In order to quantify changes in fiber organization from confocal microscopy images, a software application called FiberFit is being used to automate the objective measurement of two structural properties that describe material anisotropy: fiber orientation and fiber dispersion. The standalone software application, FiberFit, was developed in the Northwest Tissue Mechanics laboratory in 2016 and is freely available to the scientific community for Windows and Mac systems. FiberFit has now been used in many research projects outside the developing laboratory, including research on tumors, liver, muscle, and engineered composite materials. This project will re-engineer FiberFit and deploy it as a cloud-based application. Emphasis will be placed on creating a robust, intuitive, and sustainable software application that implements best practices in software engineering and design to facilitate broad adoption across the scientific community.

SPECIFIC AIMS In 2016, the Northwest Tissue Mechanics laboratory launched a standalone software application called FiberFitTM. This free software tool was developed to quantify material anisotropy in biological tissue. Material anisotropy describes the directional dependence of the material structure, and the mechanical and chemical behavior of biological tissue is regulated by the anisotropy of fibrillar networks. Two essential structural parameters that mathematically describe the material anisotropy in fibrillar soft tissue are the mean fiber orientation and the distribution of the fiber orientations, also called fiber dispersion. The accurate measurement of these parameters is of major importance in understanding structure-function relationships at molecular and macroscales in biological systems, and in the engineering of advanced materials, including textiles, electrospun scaffolds, and fiber reinforced composites. FiberFit is the first and only software application to 1) fully automate the extraction of these parameters, 2) evaluate multiple images simultaneously, and 3) provide validation of system accuracy. FiberFit software can be freely downloaded to use on Windows and Mac operating systems. In the parent R15 application, “Role of Distortion Energy in Fibroblast-Mediated Remodeling of Collagen Matrices”, FiberFit is used extensively to quantify changes in fibrillar anisotropy when applying variable cyclic loading conditions to three-dimensional cellular scaffolds. The tracked spatiotemporal changes in fibrillar alignment are then used in a growth and remodeling algorithm to describe and predict fibroblast-mediated remodeling and repair of the extracellular matrix. FiberFit has now been used in five previous publications from the PI’s lab, and has been used outside the PI’s lab, including the quantification of structural changes in tumors, liver, and muscle, and within the cell cytoskeleton. These publications showcase the extremely broad userbase that FiberFit has the potential to support. The two-dimensional images analyzed by FiberFit can be acquired from numerous imaging modalities, including light and fluorescence microscopy, SEM, TEM, SHG, and optical cameras. Length scales can vary from molecular to macroscopic, and materials can be biological, inorganic, or engineered. Despite the functional advantages and free access of FiberFit, the adoption of FiberFit into research laboratories has not achieved it’s potential. One clear barrier to the widespread adoption of FiberFit is the inherent challenges in running and maintaining standalone software on operating systems that evolve and discourage non-certified applications (e.g. Mac). This barrier can be eliminated by transitioning FiberFit to the cloud. As a cloud-based application, FiberFit could be run on most operating systems and devices, and functionality could be improved with user accounts that store images and preferences. The objective of this project is to re-engineer FiberFit to be a robust, intuitive, and sustainable cloud-based application. Aim 1: Re-engineer FiberFit to be cloud ready The python based FiberFit application will be re-engineered and be deployed as a public cloud-based application. Development of this software will follow best design practices for a robust and scalable software platform. A sequential testing framework will be utilized to ensure that the software can be easily adopted by the end users. This first aim will be completed by a senior software engineer who leads efforts in cloud deployment at Boise State University Aim 2: Promote widespread community adoption of FiberFit A successful launch of FiberFit will require supportive documentation and pre-processing features to encourage broad adoption. Documentation will include user manuals, video tutorials, embedded notifications, and helpful rollover captions. Pre-processing features will allow image cropping and reduction of edge effects. Initial deployment will follow a two-tier strategy of gradually increasing usership. To increase community awareness, a workshop featuring FiberFit will be proposed for the World Congress of Biomechanics conference. This second aim will be completed by the software engineer, a computing student, and the PI. We anticipate that this project will take 9-12 months. With a successful launch, we believe that FiberFit can impact thousands of research projects in disparate fields around the world. In the future, we envision that FiberFit will become a suite of automated and validated image processing tools that provide turn-key solutions to accelerate the pace of discovery in biomedical sciences and material science and engineering.