Page 60 - IJB-9-4

International Journal of Bioprinting Machine learning and 3D bioprinting

Table 1. Summary of ML algorithms in bioprinting
Application area Tasks ML methods Ref.
Image analysis- Identify cone mode in scaffold fabrication process in EHD CNN [14]
based in situ jetting
monitoring Identify deposit fibers’ continuity, uniformity, and regularity in Four-layer CNN, ResNeXt-50 network, linear SVM [25]
EBB classifier
Extract the flow pattern and droplet evolution in DBB Deep recurrent neural network (DRNN) [29]
Printing parameter Predict the electrospun diameter of PCL/Gt nanofibers Multiple regression, multilayer perceptron ANN [24]
optimization Identify suitable printing conditions for PPF scaffold in EBB Random forest classifiers (RFc), random forest [26]
regression (RFr)
Optimize ink composition and printing parameters in EBB SVM classifier [27]
Predict the droplet diameter in EHD inkjet printing Statistical regression analysis, GA-NN, BPNN [28]
Optimize printing parameters for GelMA and HAMA bioinks Bayesian optimization (BO) [30]
Optimize the droplet size and printing frequency in EHD inkjet Desirability function analysis [31]
Biomaterial/bioink Achieve high shape fidelity in EBB Inductive logic programming, multiple regression [32]
optimization Achieve ideal linewidth and shape fidelity in EBB Hierarchical machine learning [33]
Predict filament diameter in EBB RFr, linear regression, intrastudy linear regression [34]
Cell performance Predict cell viability SVM regression, linear regression, RFr, SVM classi- [34]
analysis fier, RFc, logistic regression classifier
Detect the impact of scaffold morphology on cell shape pheno- SVM classifier [35]
types
Analyze cell-scaffold interaction AD-GAN [36]
Predict cell-material interactions in fibrous scaffold RFr model [37]
Associate cell morphologies with diverse microenvironment SVM classifier [38]

and available dataset. Moreover, multiple methods have
been compared to identify a competitive model with better
performance [24,25] . One task can also be organized into
either a classification or regression model, and then solved
accordingly based on data processing strategies [26,30-34] .
These studies aimed to enhance the reliability and
stability of the bioprinting process as well as the mechanical
and biological performance of bioprinted constructs. In the
following section, ML methods are discussed with regard
to application areas.

2.1. Image-based in situ process monitoring
To maintain the stability and reliability of bioprinting in Figure 4. Image-based in situ process monitoring.
large-scale and long-term fabrication, the development of
in situ process monitoring is necessary. As such, the quality for identification purposes. The extrusion parameters were
of bioprinted constructs relies on intelligent printing adjusted based on model outputs. Similarly, the deposited
process control rather than operator experience.
images can be compared with predefined patterns using the
Similar to the manufacturing process, bioprinting extracted features to depict the fiber quality and pattern.
process monitoring often collects real-time extrusion and The identification model built using such features can be
deposition images, as shown in Figure 4. Subsequently, directly linked to the adjustment strategy for the deposition
image preprocessing methods are utilized to extract key parameters. Generally, extrusion and deposition images
features from the captured images, which are the inputs of [15] [25]
the corresponding ML models. These features reflect the are used to monitor EHD and EBB , respectively.
difference between the standard cone and identified cone Both traditional ML and DL methods can be used to
in EHD and deliver critical information to the ML model analyze the collected images. The former extracts image

Volume 9 Issue 4 (2023) 52 https://doi.org/10.18063/ijb.717