Page 13 - IJB-6-1
P. 13
Yu and Jiang
Figure 5. An example of using machine learning to design scaffolds for three-dimensional bioprinting.
scaffolds with different functions, the timescale 4 Conclusions
for developing biomedical or tissue engineering
coupled with bioprinting can be largely reduced Machine learning has been widely applied in
in the future. Most importantly, machine learning 3D printing for optimizing its performance
may be able to generate some unexpected novel and applications. However, few studies have
structures that can better support cell growth been reported on using machine learning in 3D
and functionalization than ever before. Figure 5 bioprinting processes. The reason for this may be
shows an example of using machine learning due to the lack of data of bioprinting as machine
(e.g., generative design method) to generate a lot learning needs enough data to do predictions and
of qualified scaffolds for being chosen or tested optimizations. While in traditional 3D printing, it
in 3D bioprinting. Once the design variables and has much more data than 3D bioprinting. Another
their corresponding range values are provided, reason is that 3D bioprinting is still new compared
the number of expected generated scaffolds can with 3D printing, and the bioprinting technique
then be set. With many options generated by itself still has many challenges to be solved.
machine learning, the scaffolds can then be tested However, we believe that bioprinting will benefit
for different objectives. a lot from machine learning in the future. In this
Another example is that machine learning can paper, a perspective on how machine learning can
be used to design better combinations of materials be used in bioprinting is proposed and illustrated.
at different concentrations for bioprinting. Specifically, machine learning can be used to
Currently, researchers have blended different gels optimize the process of bioprinting, improve or
such as collagen or hyaluronic acid to enhance analyze dimensional accuracy, defect detection,
mechanical and degradation properties. However, and material property design.
experiments on testing combinations of materials
at different concentrations are a time-consuming Acknowledgments
and expensive process. Thus, if machine learning This research is sponsored by K.C.Wong Magna
can be used to predict the temporal or structural Fund in Ningbo University.
impact of cell proliferation and extracellular
matrix deposition on the tissue construct, machine References
learning will be an effective tool to develop
or design novel biomaterials or bioprinting 1. Ng WL, Yeong WY, 2019, The Future of Skin Toxicology
techniques. Testing 3D Bioprinting Meets Microfluidics. Int J Bioprinting,
International Journal of Bioprinting (2020)–Volume 6, Issue 1 9
