White, AlisaAn estimated 1.6 million Americans live with type 1 diabetes (T1D). The most common treatment method involves daily blood glucose monitoring and insulin injections, which can negatively impact quality of life and lead to severe health complications. Whole pancreatic transplantation, a more permanent treatment option, entails invasive surgery with associated risks and a high morbidity rate. Furthermore, lifelong immunosuppressant use, which can be detrimental to health, is necessary for transplant recipients. Islet transplantation has emerged as a promising alternative for T1D treatment, offering a less invasive approach, though it still requires the use of immunosuppressants to prevent graft rejection. Encapsulation of islets in biomaterials has shown potential for mitigating immune responses post-transplantation while facilitating islet survival and insulin production. However, despite its promise, this method faces several challenges. First, a significant issue is the generation of numerous empty microcapsules during islet encapsulation, which requires an efficient method for their removal due to the limited space for the transplanted islets in patients. Second, microcapsules are typically suspended in an oil phase after generation with microfluidic devices, whereas they must be transferred into an aqueous solution for further culture or transplantation, posing technical difficulties. Third, conventional microcapsules do not provide a tissue-like environment for islets which is detrimental to islet health, and microcapsule design flaws can result in a lack of insulin production and islet cell death due to post-transplantation immune response. Furthermore, islets experience hypoxia and increased amounts of reactive oxygen species post-isolation and transplantation, resulting in islet death. This work focuses on addressing the challenges mentioned above by enhancing islet encapsulation methods through a deep learning-based on-chip detection and sorting system, enabling the creation of highly pure samples of islet-laden core-shell hydrogel microcapsules that mimic the structure and microenvironment of the pancreas. We also investigate the use of nanoparticles to encapsulate hydrophobic antioxidants for improving their delivery into islet cells to enhance islet viability after isolation and hypoxic stress. We address critical challenges in islet transplantation by investigating deep learning-enabled selective extraction, core-shell hydrogel microencapsulation, and nanoparticle-mediated antioxidant delivery. This novel multiscale biomaterials-engineering strategy has great potential for future clinical translation, contributing to the advancement of type 1 diabetes treatment.enDissertationBioengineering