Page 68 - IJB-9-1

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

exhibited a linear relationship, which can be used as a of nutrients and metabolites in the scaffold. In tissue
follow-up feedback support. The response delay of the engineering, the mechanical stiffness of the printed
pressure might cause a delay in material deposition at model plays an important role in tissue regeneration.
the path starting point and excess deposition of material When the stiffness of the printed model is greater than
at the path ending point. To avoid material deposition the present value, the concentration of stress on the
errors at start-stop points, a second pre-experiment was surrounding bone can lead to secondary bone damage. In
carried out with the target material and the optimum contrast, insufficient stiffness of the printed model may
input parameters of velocity and pressure to determine lead to implant failure and even bone atrophy. FEA is a
the degree of delay response of pressure and implement method of using mathematical approximation to simulate
the corresponding correction with a certain advance complex and real systems, which can provide a solid
start and stop of extrusion. For the common lattice theoretical basis for mechanical property evaluation. The
printing pattern, material deposition error usually FEA results indicated that the compressive modulus of
occurs at the turnarounds due to the mismatch of the the construct after feedback improved significantly and
input parameters of velocity and pressure. To avoid was closer to that of the design model. In conclusion,
material deposition errors at the turnarounds, a third our results showed that the scaffold with the “monitoring
pre-experiment was conducted by adding GCode nodes and feedback-as-you-build” mechanism was closer to the
and increasing the velocity around the turnarounds. In design model from structure to function.
contrast, Armstrong et al. avoided excessive material However, there are still some limitations to our
deposition around turnarounds by increasing the current research that need to be studied further. For
pressure and decreasing the velocity . The decreased example, only the common lattice printing path was
[5]
velocity increases the adhesion of the material to the considered in this study, and complex graded scaffold
bottom surface. However, acceleration and deceleration patterns or curved paths were not analyzed. The
[22]
strips existed around the turnarounds, and the average printing material was focused on Hap without cells
velocity was lower than the preset velocity from the inside. Future work could apply the “monitoring and
first pre-experiment. Simultaneously, considering the feedback-as-you-build” mechanism using different
response delay of pressure according to the second pre- material with different rheological properties, such as
experiment, only velocity adjustment was selected for hydrogel with cell encapsulated. In addition, the prebuilt
the turnarounds. feedback mechanism in this study relied heavily on pre-
Based on the above defect detection and the pre- experimental data with the target material, target FS, and
built feedback mechanism, a single-layer structure was LT values. The prebuilt feedback mechanism requires new
printed to verify the detection and location of the broken pre-experiments with the different nozzles and different
filament; and the second printing for defect repair. bioinks that are selected for the printing task. Large
Further, printing experiment on multi-layer scaffolds databases must be built by collecting multiple groups of
was carried out to compare the scaffolds before and after experimental data to realize fast and intelligent selection
using the presented “monitoring and feedback-as-you- of printing parameters for better feedback control. In
build” mechanism. With the “monitoring and feedback- addition, machine learning and deep learning algorithms
as-you-build” mechanism, the FS and LT values in the have gradually been applied to camera-based anomaly
straight-line path were closer to the target values, and less detection in 3D printing [23,24] , which will be introduced in
FS and LT errors occurred including start-stop points and our future work to improve the robustness and reliability
the turnarounds. In addition, the layer fidelity and overall of defect detection. In this study, the broken filament
fidelity were both higher after feedback, indicating better defect was detected and repaired with the secondary
consistency with the design model. Moreover, high- printing. However, it may encounter some other defects
fidelity printing can ensure batch consistency of printed in bioprinting, such as excess material deposition and
structures to improve the reliability of drug screening collapse. For excess material deposition, a scraper can
and ensure the degree of anastomosis between the be used to remove it in the future. For collapse during
scaffold and the defect site. In particular, for bone tissue printing, our current strategy is to detect it and terminate
scaffolds, high-fidelity printing can provide personalized the current print in time.
structure and mechanical properties that are closer to Although 3D P-OCT imaging interrupts the printing
the design model. The volume quantification results process leading to longer printing cycles, high-fidelity
showed that VP and PC improved to 46.32% and 98.78% structures is more attractive in the biomedical field. In the
after feedback from 37.68% to 98.14%, respectively, 3D P-OCT described in this study, the A-scan acquisition
which was helpful for the cell attachment and transport frequency was 50 kHz, which could be improved

Volume 9 Issue 1 (2023) 60 https://doi.org/10.18063/ijb.v9i1.624