Artificial Intelligence (AI) Based Analysis ofIn VivoPolymers and Collagen Scaffolds Inducing Vascularization

Abstract

Background/Aim: Biomaterials are essential in modern medicine, both for patients and research. Their ability to acquire and maintain functional vascularization is currently debated. The aim of this study was to evaluate the vascularization induced by two collagen-based scaffolds (with 2D and 3D structures) and one non-collagen scaffold implanted on the chick embryo chorioallantoic membrane (CAM). Materials and Methods: Classical stereomicroscopic image vascular assessment was enhanced with the IKOSA software by using two applications: the CAM assay and the Network Formation Assay, evaluating the vessel branching potential, vascular area, as well as tube length and thickness. Results: Both collagen-based scaffolds induced noninflammatory angiogenesis, but the non-collagen scaffold induced a massive inflammation followed by inflammatoryrelated angiogenesis. Vessels branching points/Region of Interest (Px^2) and Vessel branching points/Vessel total area (Px^2), increased exponentially until day 5 of the experiment certifying a sustained and continuous angiogenic process induced by 3D collagen scaffolds. Conclusion: Collagen-based scaffolds may be more suitable for neovascularization compared to non-collagen scaffolds. The present study demonstrates the potential of the CAM model in combination with AI-based software for the evaluation of vascularization in biomaterials. This approach could help to reduce and replace animal experimentation in the pre-screening of biomaterials.Biomaterials assume a pivotal role in the design and development of an extensive array of biomedical products and devices, encompassing, among others, prosthetic heart valves, hip joint replacements, dental implants, and ocular contact lenses. Of particular note is a substantial subset of 620