Novel 3D bioprinting of biomaterials : application of statistical modeling & machine learning

Access full-text files



Journal Title

Journal ISSN

Volume Title



3D bioprinting, a versatile biofabrication technique, has been widely used in various biomedical research fields. Statistical modeling and machine learning are powerful tools that can expedite the development of pharmaceutical and biomedical research. Within the scope of this dissertation, we applied statistical modeling and machine learning to different fields of bioprinting research. In Chapter 1, we reviewed the latest accomplishments in 3D printed drug delivery devices as well as the major challenges and future perspectives for additive manufacturing-enabled dosage forms and drug delivery systems. In Chapter 2, we provided a comprehensive analysis of 3D bioprinting process parameters that affect bioink printability and cell performance. We further analyzed how these parameters could be tailored to achieve the optimal printing resolution and cell performance. In Chapter 3, a combination of emulsion evaporation and extrusion-based bioprinting technique was employed to formulate polymeric microparticles. We also developed a systemic approach to assess the formulation factor significance and predict drug loading efficiency using comprehensive statistical analysis and machine learning modeling. In Chapter 4, we developed a stepwise approach to evaluate hydrogel printability qualitatively and quantitatively and employed machine learning modelling to predict ink printability. This systemic methodology demonstrates great promises in designing and predicting the properties of newly developed bioinks, expanding the potential of machine learning in biomedical fields. In Chapter 5, we performed the first global bibliometric analysis of the literature on bacteria-mediated cancer therapy from 2012 to 2021. This study provided critical insights into the historical development of this field from 2012 to 2021, which will be helpful for scientists to conduct further investigation into bacteria-mediated cancer therapy. In Appendix A, we provided an overview of the primary routes of bacteria administration for cancer treatment and discussed the advantages as well as limitations of each route. We also highlighted the application prospect of 3D bioprinting in cancer bacteriotherapy, which represents a new paradigm for personalized cancer treatment.


LCSH Subject Headings