Page 67 - IJB-9-1

International Journal of Bioprinting In situ defect detection and feedback control with P-OCT

of Layer 7 after feedback control and broken filament repair and quantified for defect detection. The feedback control
is shown in Figure 8G1. The corresponding FS and LT mechanism was built in advance for different segments
distributions are shown in Figure 8G3 and 4, respectively, of the printing path and different defects. Finally, the
and no significant FS and LT defects were observed. By effectiveness of the “monitoring and feedback-as-you-
comparing 3D P-OCT data and the design model of Layer build” quality assurance mechanism was verified by
7, the layer fidelity was calculated as 0.937 as displayed in printing with multi-layer lattice bone scaffold.
Figure 8G2, with a 10.01% improvement from 0.852 before With large-field imaging enabled by 3D P-OCT,
defect detection. Furthermore, the fidelity values of each the imaging and evaluation of the current layer can
layer in Figure 8D were calculated and are displayed in be implemented regardless of the size of the printed
Figure 8G5. The average layer fidelity after feedback control structure. FS and LT analyses, defect detection, and layer
and defect repair was significantly improved to 0.961 ± fidelity analysis can be implemented for timely feedback
0.017 (Figure 8G5) from 0.832 ± 0.024 (Figure 8G5) before control. With large-field full-depth imaging after printing,
feedback, which was comparable to the overall fidelity from the volume parameters can be analyzed, including the VP,
0.847 (Figure 8C1) and 0.931 (Figure 8D1).
PC, and fidelity of the overall printed structure.
As shown in Figure 9, 3D P-OCT data enabled
mechanical analysis and 3D structural analysis of the With 3D P-OCT data, FS and LT quantitative analyses
overall construct. After feedback, the VP and PC of the can be implemented, including the spatial distribution
construct increased from 37.68% and 98.14% to 46.32% of FS and LT defects and the detection and location of
and 98.78%, respectively (Figure 9D). Furthermore, 3D broken filaments. Furthermore, the input parameters
can be adjusted based on in situ defect detection and the
P-OCT data of the printed construct can be converted to pre-built feedback control mechanism. In the previous
STL format files using MIMICS. The mechanical stiffness
of the printed constructs before and after feedback can be work, FS was analyzed using the 2D projection images
compared with the designed model using finite element from 3D P-OCT and Euclidean distance transformation.
analysis (FEA), which was implemented to simulate the Armstrong et al. quantified filament width using surface
[18]
stress and strain process of constructs under compression points with a laser displacement scanner . Both of
using ANSYS Workbench 17.0 (Figure 9A-C). After the above methods are susceptible to small material
feedback, the compressive modulus of the construct deposition errors in the vertical direction of the path, such
improved from 84.4374% to 33.3622%, which was closer as small burrs, resulting in FS calculation deviation. In
to that of the design model (22.09%). this study, FS quantization was based on 3D P-OCT data
and Euclidean distance transformation in 3D space, with
4. Discussion and conclusion improved quantitative accuracy. In the previous work, LT
analysis was performed only for the height of the actual
3D bioprinting provides new technology for tissue and material deposition, ignoring the consistency between
organ regeneration, drug screening, disease modeling, the actual material deposition path, and the designed
and other fields. 3D printing technology with high- path. To quantify LT defects and detect broken filaments,
fidelity structure and function is key to promoting the the designed path was combined with LT analysis in this
large-scale application of 3D bioprinting in biomedical study. First, GCode nodes were interpolated to make the
field. However, printing defects lead to low fidelity from resolution consistent with that of the 3D P-OCT data, and
structure to function, due to the lack of in situ defect the LT was analyzed at each node to determine the LT
detection and timely feedback control. In situ defect distribution and defect detection along the printing path.
detection and location, timely feedback control, and
defect repair are necessary to promote the application Based on large-field full-depth imaging with 3D
of 3D bioprinting to accurately manufacture complex P-OCT, and FS and LT quantitative analyses, the
personalized structures. The “monitoring and feedback- feedback control mechanism can be pre-built to adjust
as-you-build” quality assurance mechanism were the input parameters and defect repair. In this study,
presented to improve printing efficiency, reduce material material deposition errors under three different paths
waste, and maximize the printed structure’s fidelity were considered for the pre-built feedback mechanism,
to the design, thus promoting the application and including start-stop points, straight-line paths, and the
promotion of 3D bioprinting in organ transplantation turnarounds. The first pre-experiment was carried out
and disease modeling. First, in situ process monitoring to explore the relationship between the target material
was achieved using 3D P-OCT for large-field full-depth and two printing parameters, velocity and pressure, and
imaging based on point cloud registration. Based on the FS and LT of the filament extruded through a nozzle.
imaging data, spatially resolved FS and LT were analyzed Under the same pressure value, the FS value and velocity

Volume 9 Issue 1 (2023) 59 https://doi.org/10.18063/ijb.v9i1.624