Investigation on the bio-ink properties in influencing printability of inkjet bio-printing

• Main Author: Bu, Marcus Jen Jack
• Other Authors: Yeong Wai Yee
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Mechanical engineering
• Online Access: https://hdl.handle.net/10356/168089
• Institution: Nanyang Technological University

The advancements in bio-printing feature the growing significance of biodegradability and humanity. Though numerous studies and research have been conducted on plant-based biomaterials to prevail on the current limitations, the investigation and development for attaining plant-based bio-ink’s printability without physical means still require further exploration. This paper aims to address this current limitation by the utilization of Thermal Inkjet Bio-printers to obtain specific velocities of the plant-based bio-ink droplet, which will be implemented to develop a desirable Machine Learning model for characterizing and predicting viable plant-based bio-inks, as well as their velocity profile, upon print. This report describes the viable solutions of plant-based bio-inks suitable for Thermal Inkjet Bio- printing. With the utilization of the rheometer, as well as various test procedures, the rheological and mechanical properties of the plant-based bio-inks were recognized. Furthermore, multi-solute plant- based bio-inks were characterized and established. Through the advancements of Thermal Inkjet Bio-printers incorporated with high-speed cameras to study the character of a plant-based bio-ink droplet, the velocity profile of the plant-based bio-ink droplets was captured and defined. In addition to the velocity profile, the Thermal Inkjet Bio-printer was employed to visually assess the plant-based bio-ink droplet printability, which was utilized to acquire the Printability Score (PS). Furthermore, the prediction of Printability, PS value, as well as velocity profile of plant-based bio-inks utilized Machine Learning models, namely Linear Regression, Decision Tree Regressor, Random Forest Regressor, Decision Tree Classifier and Logistic Regression upon data analysis. Prediction of Printability from PS values was established and proven to be extremely harmonious, with the utilization of a Decision Tree Classifier model. The predicted velocity profile was rather accurate, with two specific clusters. This proved the precision and accuracy of the predicted velocity profile. In conclusion, the Machine Learning models employed have the potential to reduce repetitive labour as well as material wastage. Additionally, velocity profiles of plant-based bio-inks can be obtained without further rigorous experimentation, presenting an array of promising prospects for future applications.