In Situ Characterization of Melt–Electrowritten Scaffolds in 3D Using Optical Coherence Tomography

Abstract

Recent developments in melt electrowriting (MEW), a high‐resolution additive manufacturing technology, have led to increases in scaffold complexity. However, MEW scaffolds are currently characterized ex situ, which causes time–consuming iterations of characterization and fabrication that limit scaffold throughput and more widespread use of the technology. For the first time, an in situ method to characterize the 3D microstructure of MEW scaffolds using optical coherence tomography (OCT) is described. Calculations of microstructural features are performed on OCT data using a custom algorithm, demonstrating close correspondence with scanning electron microscopy (SEM). For example, OCT calculations of fiber diameter and scaffold thickness are within an average of 0.31 and 1.79 μm, respectively, of corresponding SEM–derived calculations. Additionally, the 3D capabilities of OCT enable the nondestructive characterization of scaffolds with depth–varying microstructures, overcoming some main limitations of SEM. Finally, in situ characterization is achieved by integrating the OCT scanner within an MEW printer, enabling the scaffold microstructure to be evaluated and optimized during manufacture. This new capability represents an important step toward achieving an efficient fabrication–characterization cycle with the guaranteed scaffold quality and reproducibility required to validate the manufacturing process.

Introduction

Melt electrowriting (MEW) is a high-resolution additive manufacturing technology often used in tissue engineering to print fibrous scaffolds that guide cellular attachment and proliferation.^[] Recent breakthroughs in MEW have enabled scaffold designs with highly controlled 3D microstructures, making it attractive for a wide range of applications including heart patches,^[] tympanic membrane replacements,^[] heart valves,^[] and bone grafts.^[] The complexity of MEW scaffolds has diversified with multiple pore sizes,^[] fiber diameters,^[] and other 3D features being incorporated into a single scaffold design to recapitulate regional differences in tissue microstructure. However, the current method for achieving accurate and reproducible microstructures relies on the user's expertise to adjust input printing parameters, such as applied voltage, air pressure, and collector speed.^[] We hypothesize that in situ 3D characterization of MEW scaffolds could reduce the reliance on the user's expertise and provide a means to ensure high quality.

Currently, scanning electron microscopy (SEM) and stereomicroscopy are the most widely used techniques to characterize MEW scaffolds.^[] SEM provides excellent resolution on the nanometer scale; however, it is a destructive technique with no depth–sectioning capabilities.^[] Alternatively, stereomicroscopy is a nondestructive and rapid method to visualize samples post-manufacture on the microscale; however, it is also restricted to 2D imaging. As such, these two techniques are limited in their ability to characterize complex scaffold designs with depth–varying microstructural properties.

The limitations of SEM and stereomicroscopy have led to the exploration of X–ray microcomputed tomography (μCT) for 3D scaffold characterization.^[] Youssef et al. demonstrated the capability of μCT to provide 3D reconstructions of scaffolds with resolutions ranging from 0.5 to 3 μm, enabling volume porosity and scaffold thickness to be quantified,^[] for example, in aortic root scaffolds.^[] Mineralization in scaffolds cultured with human osteoblasts^[] has also been achieved with μCT. However, μCT is expensive and must be housed in a relatively large space, and typically requires several hours to image a single sample.^[] While μCT and SEM both provide high–resolution imaging capabilities, they are time–consuming processes that require dedicated facilities, resulting in an inherently low throughput. As a result, the current processes for characterizing MEW scaffolds are laborious iterative cycles between fabrication and characterization that represent a key quality control barrier to eventual clinical use. In situ 3D characterization and monitoring of MEW scaffolds during manufacture would address these issues to improve scaffold quality. Such validation could include calculating essential scaffold features, such as fiber diameter, pore size, scaffold thickness, and fiber stacking.^[] In situ monitoring of the MEW jet Taylor cone and flight path using machine vision to monitor the print process has been investigated to assess printing stability.^[] This technique does not visualize the scaffold 3D microstructure and so cannot provide feedback on scaffold quality.

The manufacturing challenges outlined earlier are addressed here using optical coherence tomography (OCT). OCT is a high–resolution optical imaging technique that has the potential to combine accurate 3D characterization of MEW scaffolds with in situ monitoring. OCT utilizes low-coherence interferometry, with 3D images generated based on backscattered light from refractive index boundaries in a sample.^[] OCT provides depth sectioning up to 1–2 mm with spatial resolution of 1–10 μm,^[] and is a clinical standard in ophthalmology for diagnosis of retinal diseases.^[] Furthermore, OCT is emerging as an important clinical tool in areas such as cardiology^[] and dermatology,^[] as well as for use in tissue engineering.^[] OCT has been used to measure porosity and cell integration in engineered bone tissue,^[] to image tissue–engineered skin,^[] and to monitor the evolution of collagen fiber alignment in tissue–engineered tendons.^[] Imaging over centimeter fields of view can be achieved in several seconds,^[] potentially enabling real-time 3D characterization of scaffolds when integrated into MEW printers. Additionally, OCT is substantially lower in cost than either μCT or SEM, with commercial systems now available for < 10 000 USD,^[] and, also, OCT scanners are readily implemented in compact formats.^[]

Here, for the first time, we demonstrate accurate, nondestructive, 3D characterization of MEW scaffold features in situ using OCT. We demonstrate OCT-derived calculations of fiber diameter and scaffold thickness to be within 0.31 ± 1.06 and 1.79 ± 0.50 μm, respectively, of corresponding SEM–derived calculations. We additionally demonstrate the characterization of depth–varying pore size, and evaluation of fiber diameter and scaffold microstructure in situ by integrating OCT into an MEW printer. Our results demonstrate that OCT is suitable for the rigorous evaluation of MEW scaffolds that is required for in situ, stringent, quality control of products in clinical applications.

Results and Discussion

3D Microstructural Characterization of Scaffolds with OCT

To demonstrate the suitability of OCT to characterize MEW scaffolds, we first developed an image analysis algorithm in MATLAB (Mathworks, USA) to measure fiber diameter, scaffold thickness, and pore size from 3D images that were acquired ex situ. Ten scaffolds (Scaffolds 1–10) were designed, printed, and used in this study (more details are provided in Table S1, Supporting Information). Figure  shows how these scaffold features were measured, using a 2 × 2 mm section of Scaffold 1, which comprised three distinct square–shaped pore sizes that changed with depth every two layers. A 3D image of the scaffold positioned on a glass slide was acquired with OCT (Figure ). As shown in Movie S1, Supporting Information, 3D–OCT reveals the depth–varying microstructure of the scaffold, providing critical insight to the overall print quality, which is otherwise difficult to obtain with 2D SEM images (see Figure S1a, Supporting Information).

[IMAGE OMITTED. SEE PDF]

Fiber diameter and scaffold thickness were assessed using a set of cross–sectional OCT images (B–scans), which were averaged over a spatial range of 13–22 μm to reduce noise. Figure  is an averaged B–scan showing a section of the scaffold wall containing six layers. There are seven regions of high OCT intensity in the B-scan, that is, the white horizontal lines, which result from the change in refractive index at the interface between the edges of the fibers and air (see Supporting Information and Figure S2, Supporting Information). Fiber diameter was calculated in the z–direction by measuring the distance between these lines, achieved by plotting the signal intensity with depth at a specified location. This resulted in an intensity profile (A-scan) (Figure ), and the distance between adjacent peaks corresponds to the air–fiber interfaces. Similarly, scaffold thickness was computed by measuring the distance between the first and last peaks of the intensity profile (Figure ). Next, pore size was measured from maximum intensity projections of a set of en face images, which are images in the x–y plane (Figure ). Maximum intensity projections were used to account for topographical changes in fiber deposition. Intensity profiles were plotted in the x- and y- directions where the distance between each peak was used to compute the x- and y-dimensions of the pore (Figure ). To further increase the precision of the fiber diameter, scaffold thickness, and pore size measurements, a Gaussian kernel was convolved with the raw signal to reduce the impact of noise and to improve peak localization (see Supporting Information).

Evaluation of Fiber Diameter and Scaffold Thickness in 3D Using OCT

The accuracy of fiber diameter and scaffold thickness, as measured by OCT, was evaluated by validating measurements taken from three scaffolds (Scaffolds 2, 3, and 4) against SEM images from the same spatial location (Figure ). Scaffolds 2 and 3 were designed with square–shaped pores and three distinct fiber diameters that changed every two layers (an SEM image of Scaffold 2 is shown in Figure S1b, Supporting Information). These scaffolds were compared with Scaffold 4, which was designed with a constant fiber diameter. OCT fiber diameter and scaffold thickness was calculated as previously described from a total of five regions: two from Scaffold 2, two from Scaffold 3, and one from Scaffold 4. Each region was composed of an averaged B–scan spanning a 450 μm section of the scaffold wall (a region from Scaffold 2 is shown in Figure ). After OCT imaging, the corresponding regions of scaffold wall were cut with a surgical blade, as illustrated in Figure , and imaged with SEM (an example from Scaffold 2 is shown in Figure ). As illustrated in Figure , OCT measurements were found to correspond closely to SEM measurements. Bar charts comparing fiber diameter measurements for the other four regions are displayed in Figure S3, Supporting Information.

[IMAGE OMITTED. SEE PDF]

The accuracy of OCT measurements was quantified with respect to corresponding SEM measurements by calculating the difference between the mean measurement for every fiber diameter and scaffold thickness (mean OCT measurement—mean SEM measurement). As shown in the vertical scatter plot in Figure , the range of accuracy for all fibers across all regions was −0.31 ± 1.06 μm, showing very close correspondence with SEM. This high accuracy demonstrates that OCT measurements can be nondestructively obtained for fibers throughout the scaffold depth, providing valuable information on the printing stability and scaffold quality. Similarly, the range of accuracy for scaffold thickness was −1.79 ± 0.50 μm (Figure ). This highlights the capability of OCT to measure scaffold features that cannot be obtained nondestructively with 2D characterization techniques like SEM. Some possible contributions to the minor difference between OCT and SEM include the resolution of the OCT system (5 μm), the presence of noise, and signal attenuation with depth in the OCT images (depicted in Figure ). The limitation of resolution could be addressed by implementing optical coherence microscopy, a high-resolution variant of OCT, that provides sub-micrometer resolution.^[] More sophisticated OCT signal processing and the use of a Bessel beam to increase the depth of field^[] may also improve the precision of these measurements.

Evaluation of Pore Size Using OCT

Analogously to fiber diameter and scaffold thickness measurements, the accuracy of OCT pore size calculations was evaluated against corresponding SEM images using two scaffolds (Scaffolds 5 and 6). Three pores were used for validation, where one pore was taken from Scaffold 5, which was designed with 2 × 2 mm square–shaped pores (Figure S1c, Supporting Information), and the remaining two were taken from Scaffold 6 with 0.75 × 0.75 mm pores. Ten OCT measurements were taken across the x-dimension over the central third of the pore, and the same process was repeated in the y-dimension. The purple rectangle with the solid line outlines the central third of the pore where the x–dimension measurements were taken from Scaffold 5 and the purple rectangle with the dotted line shows the location of the y–dimension measurements (Figure ). The green rectangles in Figure  outline the corresponding location in an SEM image of Scaffold 5. The dimensions of each pore were defined to begin and end at the center of the fibers making up the pore edges (shown by the arrows in Figure ). Similar to fiber diameter and scaffold thickness, the accuracy of OCT measurements was quantified as the difference between the mean of the OCT measurements and the corresponding mean of the SEM measurements for the x- and y-dimensions (see Section ). The bar chart in Figure  shows close correlation between OCT and SEM measurements in the x- and y-dimensions of the pore from Scaffold 5. The range of accuracy for all pore dimensions across the three pores was −6.98 ± 16.82 μm, demonstrating the suitability of OCT in characterizing pore size for MEW scaffolds. This accuracy is not as high as the results for fiber diameter and scaffold thickness due to unavoidable handling of the scaffolds as they were prepared for SEM imaging, causing distortions in pore shape that alter the pore size measurements.

[IMAGE OMITTED. SEE PDF]

We next explored the capability of OCT to characterize pore size in 3D, that is, pore sizes that vary with depth. Two scaffolds (Scaffolds 1 and 7) were printed, and imaged with OCT (Figure  and Figure S1a,d, Supporting Information). Scaffold 1 (Figure ) was designed, from the top–down, to have three phases: three layers of 2 × 2 mm square pores (Phase 1), three layers of 1 × 1 mm square pores (Phase 2), and three layers of 0.5 × 0.5 mm square pores (Phase 3). Scaffold 7 (Figure ) was also designed with three phases: four layers of 2 × 2 mm square pores (Phase 1), a melt electrospun mesh with a 1 mm arc radius and 75 cycles (Phase 2), and six layers of 0.5 × 0.5 mm square pores (Phase 3). 3D-OCT enabled the three phases in each scaffold to be segmented from the images via maximum intensity projections so that pore size could be measured throughout the entire depth of the scaffold, as shown in Figure . 3D animations of Scaffolds 1 and 7 are shown in Movie S1,S2, Supporting Information.

[IMAGE OMITTED. SEE PDF]

To obtain quantitative information, a pore from each phase was randomly selected and measured using the process described previously. The x- and y-pore dimensions for each phase are presented in Figure , where the measurements are compared to the preprogrammed target values for each phase (shown by the horizontal black dotted lines). By comparing the output pore size to the target values, the discrepancy between the programmed stage movement and the deposited fiber pattern was assessed.^[] For example, Figure  shows that the pore size for each of the three phases in Scaffold 1 were close to the target inputs, indicating a stable printing process and a small jet lag. Similarly, the pore size of Phase 3 for Scaffold 7 was also aligned with the target input. However, the y-dimension of Phase 1 in Scaffold 7 was 270 μm (13%) short of the target input (Figure ). This error was likely caused not only by the jet lag but also by the deposition of this layer over an irregular surface caused by the underlying layer of electrospun fibers (Phase 2). These results demonstrate that an accurate and quantitative comparison of the pore size measurements to the preprogrammed laydown pattern can be obtained with OCT images throughout the scaffold depth, overcoming the 2D limitation of SEM images by providing feedback on the 3D microstructure to enable the detection of defects and correction of the lag. This is particularly beneficial for Scaffold 7 (Figure ), where the electrospun mesh (Phase 2) disguises the pores in Phase 3 underneath (shown by the SEM image in Figure S1d, Supporting Information).

Similar to the design of Scaffold 7, the utilization of a thin membrane mesh phase that separates two different phases has recently been explored in MEW designs.^[] Such scaffolds have utility in tissue boundary applications such as treating periodontal disease,^[] where separate tissues require different morphologies, and for the purpose of improving cell–seeding efficiency.^[] One key challenge with these designs is that the dense membrane obscures the underlying phase and thus is impossible to visualize with a single SEM image, without cutting the sample. As such, we used OCT to further test its capability in characterizing pore size by imaging scaffolds with a dense membrane phase, as previously described.^[] Animations of the 3D-OCT reconstructions of these scaffolds are shown in Movies S3,S4, Supporting Information, which are shown to reveal the underlying phase disguised by the dense mesh phase. In these animations, the 3D microstructure of the scaffold can be clearly interpreted, enabling valuable information on scaffold quality to be obtained efficiently and nondestructively.

In Situ Monitoring of Scaffold Features Using OCT

Next, we investigated whether OCT characterization of MEW scaffolds could be achieved in situ, that is, during manufacture. We demonstrated this in three steps: by designing an integrated OCT-MEW system (Figure ), by measuring fiber diameter in situ (Figure ), and by monitoring scaffold microstructure such as the effect of layer offsetting in situ (Figure ). As shown by the model of the OCT–MEW system (Figure ), the OCT scanner was positioned above the collector plate and was fixed to the side panel of the MEW printer via a 3D printed bracket. To image a scaffold, the collector was translated such that the scaffold was positioned under the OCT beam. The scanner was also attached to a translation stage so the depth of focus could be adjusted prior to printing.

[IMAGE OMITTED. SEE PDF]

To measure the fiber diameter in situ, we printed Scaffold 8, using the same design as Scaffold 6 (Figure ), and set an arbitrary target diameter of 40 ± 2 μm. Using a systematic fiber calibration process shown by the schematic in Figure , we acquired OCT images of fibers that were printed after stabilizing the jet. From these OCT images, three measurements of fiber diameter were taken in situ using the OCT system manufacturer's software (ThorImage, ThorLabs, USA), where we manually measured the fiber diameter in the B–scan (Figure S4, Supporting Information). Importantly, this simple calibration step enabled us to rapidly adjust the input printing parameters to achieve the target fiber diameter under stable conditions. Figure  shows 3D reconstructions of three fibers that were printed under different pressure and collector speeds to tune the fiber diameter. At 1.5 bar and 400 mm min⁻¹, the fiber diameter was 18.8 μm, which was half the target diameter (Figure ). The pressure was then increased to 2.2 bar and the collector speed reduced to 200 mm min⁻¹, which produced a 37.5 μm diameter which was near the target range; however, coiling occurred (Figure ). By increasing the collector speed to 330 mm min⁻¹, the coiling effect was removed and a stable fiber with a diameter of 38.5 μm was achieved, which was in the target range (Figure ). This in situ measurement was validated against ex situ OCT measurements of fiber diameter from the printed scaffold. A total of 765 measurements were taken from four arbitrarily selected regions that were 100 μm in length. Ex situ measurements of fiber diameter were found to be 37.5 ± 2.1 μm, confirming the in situ measurement to be within one standard deviation. These measurements validated the suitability of OCT as an in situ technique to tailor the fiber diameter prior to starting a new print and to monitor the stability of the printing process. In future, our image analysis algorithm could be incorporated with in situ monitoring to automate this process.

Performing in situ characterization of scaffold microstructure and layer offsetting is particularly important for sinusoidal scaffold designs where the scaffold walls tend to tilt in the direction of writing.^[] This tilting of the fiber walls is known to occur when printing above the critical translation speed and with a tool path that is changing direction. This is especially pertinent for sinusoidal scaffolds such as those in Figure . The tilting of the walls was recently used to control the stretching mechanics of MEW scaffolds.^[] Using an approach termed microscale layer shifting, small changes in the tool path can create defined fiber wall tilting, overhangs, textured patterns, and more complex geometries. OCT is compatible with measuring fiber wall tilting, providing in situ information on their formation rather than using the standard (and destructive) secondary step of SEM.

With our integrated OCT–MEW system, we aimed to visualize the scaffolds in situ to estimate the amount of layer offset required to ensure vertical stacking of fibers. A single–phase sinusoidal scaffold was designed with five layers and 2 mm auxetic bowtie pores (Scaffold 9). In this design, no layer shifting was used, resulting in inward tilting of walls as shown in the 3D-OCT reconstruction of Scaffold 9 (Figure ), which was taken in situ. The B-scans of the scaffold also confirmed this inward tilting of the fiber stacks (Figure ). Using this information, the scaffold was printed again, this time using layer shifting with a layer offset of 10 μm (Scaffold 10). As shown in the 3D-OCT reconstruction of this scaffold (Figure ), the fibers were stacked vertically. The B-scans of the scaffold also confirmed improved fiber stacking (Figure ). This analysis demonstrates that in situ 3D-OCT reconstructions provide valuable information on both fiber stacking and microstructure, which in the future could be used to correct these geometrical features in quasi–real time to improve scaffold quality and reproducibility.

Conclusion

For the first time, an in situ characterization method for MEW scaffolds was developed using OCT that enabled 3D microstructural characteristics to be quantified nondestructively during scaffold manufacture. The capability to characterize MEW scaffolds with OCT was first demonstrated via ex situ calculations of fiber diameter, scaffold thickness, and pore size, which were validated against SEM on scaffolds with depth–varying properties. The depth–sectioning capabilities of OCT overcome the 2D limitations of SEM, enabling a rigorous evaluation of the 3D microstructure and detection of defects. In this study, OCT was integrated within an MEW printer, enabling in–process feedback on fiber diameter and scaffold microstructure so that the input printing parameters could be manually adjusted to fabricate the target scaffold. These novel characterization methods could be extended to characterize other volumetric scaffold features such as porosity, layer fusion, and layer stacking. While in this study, in situ characterization was implemented manually, in future, this process could be automated to achieve a closed loop feedback system within MEW printers. This would allow nonexpert users to fabricate high-quality scaffolds without the fabrication–characterization cycle. This holds great promise for the upscaling of MEW, and its translation to manufacturing and routine use in clinical applications.

4. Experimental Section

Manufacturing of Poly(ε-caprolactone) Scaffolds

The ten scaffolds in this study were manufactured using medical grade poly(ε-caprolactone) (PCL) (Purasorb PC 12, Corbion, The Netherlands). This polymer is commonly used in tissue engineering applications for its biocompatibility, slow biodegradability, low melting point (60 °C), and high thermal stability.^[] The scaffolds were printed with a MEW printer that was built in–house where the PCL pellets were heated in a plastic syringe (Nordson EFD, USA) at 75 °C. Using air pressure, the PCL melt was extruded through a 23 G nozzle heated at 85 °C, while the collector distance was kept constant throughout each individual print. A voltage was used to draw the polymer melt into a thin jet onto a 1 mm thick glass microscope slide, which laid atop a laterally translating aluminum collector. Printing onto glass slides enabled simple transfer of the scaffold to OCT for imaging. The printing parameters used for all scaffolds in this study are detailed in Table S1, Supporting Information.

OCT Image Acquisition

All MEW scaffolds were imaged using a benchtop spectral–domain OCT system (Telesto III OCTG, Thorlabs, USA). The center wavelength of the light source was 1300 nm and the 3 dB bandwidth was 170 nm. The measured full width at half maximum (FWHM) axial resolution in air is 4.8 μm. The system was equipped with a mechanically adjustable stage and an objective lens (LSM02, ThorLabs, USA), providing a lateral field of view of 4.2 × 4.2 mm and a lateral resolution of 4.4 μm in air, measured as the FWHM of the spot size of the beam. The system utilized a 12–bit line camera detector and had a dynamic range of 60 dB. The scaffolds were transported on the glass slides and positioned under the OCT system with the fibers aligned with the scanning direction of the OCT beam. The location of the focus was set at the axial center of the scaffold to maximize the amount of scaffold within the depth of field. The reference arm power and exposure time were adjusted to maximize the image intensity without saturating the detector. 3D images were then acquired using both ThorImage (ThorLabs, USA) for creating 3D reconstructions and a custom-made software, which enabled importing of the data into MATLAB (Mathworks, USA) to measure scaffold features. The OCT beam was scanned in a raster square at an A–scan acquisition rate of 50 kHz and exposure time of 10 μs. A 2.2 × 2.2 mm field of view was used for all scaffolds except Scaffold 7, which was scanned over a 1.8 × 1.8 mm field of view. To ensure that OCT images were acquired at the optical resolution of the OCT system, in the x- and y-directions, Nyquist sampling was achieved by ensuring that the pixel size was at least 3 times smaller than the OCT system resolution.^[] To increase the signal–to–noise ratio (SNR) of OCT scans, five B-scans were averaged at each y-location. The SNR is increased by this temporal averaging as the optical noise realization is different each time the B-scan is acquired over a specific location. The OCT data was processed using MATLAB and custom–made signal processing code to convert the raw spectral data into SNR values. More information on OCT signal processing and noise removal is detailed in the Supporting Information and Figure S5, Supporting Information.

SEM Image Acquisition and Processing

To validate OCT measurements of fiber diameter, thickness, and pore size, scaffolds were imaged under SEM. The scaffolds were first mounted onto metallic pins using double–sided conductive carbon tape and sputter coated with a ≈3 nm layer of platinum using a Polaron SC7640 (Quoram Technologies, UK). Images were acquired using a Zeiss Gemini SEM (Carl Zeiss Microscopy, Germany) with a 30 μm aperture, 5 kV accelerating voltage, and a working distance of 11.6–12.1 mm. For Scaffolds 2 and 3, the sputter coating was too fine due to a fault in the machine, so a smaller aperture (10 μm), higher accelerating voltage (10 kV), and larger working distance (21.4 mm) were used. SEM measurements were taken at the same spatial location of OCT measurements, using the measuring tool on ImageJ (National Institutes of Health, USA).

Statistical Analysis

All fiber diameter, scaffold thickness, pore size, and measurement accuracies were reported as mean ± SD. Any values that were greater than two SDs away from the mean were considered statistical outliers and excluded from the data. Unpaired T-tests were used to compare measurements between OCT and SEM. Values of p < 0.05 were considered significant and the p < 0.05* was used to indicate significance in all bar charts.

Acknowledgements

E. M. D.-J.-P. and B. F. K., equally contributed to this work. The authors acknowledge the facilities, and the scientific and technical assistance of Microscopy Australia at the Centre for Microscopy, Characterisation & Analysis, The University of Western Australia, a facility funded by the University, State and Commonwealth Governments. The authors also acknowledge Dr. Andrei Hrynevich for the fabrication of the scaffolds in Movies S3 and S4, Supporting Information.

Conflict of Interest

The authors declare no conflict of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

N. T. Saidy, F. Wolf, O. Bas, H. Keijdener, D. W. Hutmacher, P. Mela, E. M. De-Juan-Pardo, Small 2019, 15, 1900873.

I. Liashenko, A. Hrynevich, P. D. Dalton, Adv. Mater. 2020, 32, 2001874.

F. M. Wunner, M.‐L. Wille, T. G. Noonan, O. Bas, P. D. Dalton, E. M. De-Juan-Pardo, D. W. Hutmacher, Adv. Mater. 2018, 30, 1706570.

M. Castilho, A. van Mil, M. Maher, C. H. G. Metz, G. Hochleitner, J. Groll, P. A. Doevendans, K. Ito, J. P. G. Sluijter, J. Malda, Adv. Funct. Mater. 2018, 28, 1803151.

M. von Witzleben, T. Stoppe, T. Ahlfeld, A. Bernhardt, M.‐L. Polk, M. Bornitz, M. Neudert, M. Gelinsky, Adv. Healthc. Mater. 2021, 10, 2002089.

N. Abbasi, S. Ivanovski, K. Gulati, R. M. Love, S. Hamlet, Biomater. Res. 2020, 24, 2.

A. Hrynevich, B. Ş. Elçi, J. N. Haigh, R. McMaster, A. Youssef, C. Blum, T. Blunk, G. Hochleitner, J. Groll, P. D. Dalton, Small 2018, 14.

F. M. Wunner, P. Mieszczanek, O. Bas, S. Eggert, J. Maartens, P. D. Dalton, E. M. De-Juan-Pardo, D. W. Hutmacher, Biofabrication 2019, 11, 025004.

F. M. Wunner, O. Bas, N. T. Saidy, P. D. Dalton, E. M. D.‐J. Pardo, D. W. Hutmacher, J. Vis. Exp. JoVE 2017, 130, 56289.

A. Youssef, A. Hrynevich, L. Fladeland, A. Balles, J. Groll, P. D. Dalton, S. Zabler, Tissue Eng. Part C Methods 2019, 25, 367.

O. Bas, D. D'Angella, J. G. Baldwin, N. J. Castro, F. M. Wunner, N. T. Saidy, S. Kollmannsberger, A. Reali, E. Rank, E. M. De-Juan-Pardo, D. W. Hutmacher, ACS Appl. Mater. Interfaces 2017, 9, 29430.

J. Kim, E. Bakirci, K. L. O'Neill, A. Hrynevich, P. D. Dalton, Macromol. Mater. Eng. 2021, 306, 2000685.

O. Bas, E. M. De-Juan-Pardo, C. Meinert, D. D'Angella, J. G. Baldwin, L. J. Bray, R. M. Wellard, S. Kollmannsberger, E. Rank, C. Werner, T. J. Klein, I. Catelas, D. W. Hutmacher, Biofabrication 2017, 9, 025014.

P. Mieszczanek, T. M. Robinson, P. D. Dalton, D. W. Hutmacher, Adv. Mater. 2021, 33, 2100519.

S. T. Ho, D. W. Hutmacher, Biomaterials 2006, 27, 1362.

N. T. Saidy, T. Shabab, O. Bas, D. M. Rojas-González, M. Menne, T. Henry, D. W. Hutmacher, P. Mela, E. M. De-Juan-Pardo, Front. Bioeng. Biotechnol. 2020, 8, 793.

C. Fella, A. Balles, R. Hanke, A. Last, S. Zabler, Rev. Sci. Instrum. 2017, 88, 123702.

A. du Plessis, C. Broeckhoven, Acta Biomater. 2019, 85, 27.

A. C. Jones, B. Milthorpe, H. Averdunk, A. Limaye, T. J. Senden, A. Sakellariou, A. P. Sheppard, R. M. Sok, M. A. Knackstedt, A. Brandwood, D. Rohner, D. W. Hutmacher, Biomaterials 2004, 25, 4947.

M. Bartos, T. Suchy, Z. Tonar, R. Foltán, M. Kalbacova, Ceram. - Silik. 2018, 62, 1.

A. Gh, J. Microsc. 2012, 247, 209.

Optical Coherence Tomography: Technology And Applications (Eds: W. Drexler, J. G. Fujimoto), Springer International Publishing 2015.

D. P. Popescu, L.‐P. Choo-Smith, C. Flueraru, Y. Mao, S. Chang, J. Disano, S. Sherif, M. G. Sowa, Biophys. Rev. 2011, 3, 155.

K. Grieve, M. Paques, A. Dubois, J. Sahel, C. Boccara, J.‐F. L. Gargasson, Invest. Ophthalmol. Vis. Sci. 2004, 45, 4126.

W. Drexler, U. Morgner, R. K. Ghanta, F. X. Kärtner, J. S. Schuman, J. G. Fujimoto, Nat. Med. 2001, 7, 502.

N. Nassif, B. Cense, B. H. Park, S. H. Yun, T. C. Chen, B. E. Bouma, G. J. Tearney, J. F. de Boer, Opt. Lett. 2004, 29, 480.

F. Alfonso, M. Paulo, N. Gonzalo, J. Dutary, J.‐Q. Pillar, V Lennie, J. Escaned, C Bañuelos, R. Hernandez, C. Macaya, J. Am. Coll. Cardiol. 2012, 59, 1073.

I.‐K. Jang, B. E. Bouma, D.‐H. Kang, S.‐J. Park, S.‐W. Park, K.‐B. Seung, K.‐B. Choi, M. Shishkov, K. Schlendorf, E. Pomerantsev, S. L. Houser, H.T. Aretz, G. J. Tearney, J. Am. Coll. Cardiol. 2002, 39, 604.

P. Barlis, J. Schmitt, EuroIntervention 2009, 4, 529.

T. C. M. P. Blumetti, M. P. Cohen, E. E. Gomes, M. P. de Macedo, J. Am. Acad. Dermatol. 2015, 73, 315.

Z. Wang, H. Pan, Z. Yuan, J. Liu, W. Chen, Y. Pan, Tissue Eng. Part C Methods 2008, 14, 35.

J. Welzel, Skin Res. Technol. 2001, 7, 1.

M. J. J. Liu, S. M. Chou, C. K. Chua, B. C. M. Tay, B. K. Ng, Med. Eng. Phys. 2013, 35, 253.

K. Zheng, M. A. Rupnick, B. Liu, M. E. Brezinski, Open Tissue Eng. Regen. Med. J. 2009, 2, 8.

Y. Yang, A. Dubois, X. Qin, J. Li, A. E. Haj, R. K. Wang, Phys. Med. Biol. 2006, 51, 1649.

L. E. Smith, M. Bonesi, R. Smallwood, S. J. Matcher, S. MacNeil, J. Tissue Eng. Regen. Med. 2010, 4, 652.

M. Ahearne, P. O. Bagnaninchi, Y. Yang, A. J. E. Haj, J. Tissue Eng. Regen. Med. 2008, 2, 521.

G. Guan, M. Hirsch, Z. H. Lu, D. T. D. Childs, S. J. Matcher, R. Goodridge, K. M. Groom, A. T. Clare, Mater. Des. 2015, 88, 837.

J. Allen, Lumedica n.d.

M. J. Gora, M. J. Suter, G. J. Tearney, X. Li, Biomed. Opt. Express 2017, 8, 2405.

H. Pahlevaninezhad, M. Khorasaninejad, Y.‐W. Huang, Z. Shi, L. P. Hariri, D. C. Adams, V. Ding, A. Zhu, C.‐W. Qiu, F. Capasso, M. J. Suter, Nat. Photonics 2018, 12, 540.

R. A. Leitgeb, Biomed. Opt. Express 2019, 10, 2177.

R. A. Leitgeb, M. Villiger, A. H. Bachmann, L. Steinmann, T. Lasser, Opt. Lett. 2006, 31, 2450.

A. Hrynevich, I. Liashenko, P. D. Dalton, Adv. Mater. Technol. 2020, 5, 2000772.

C. Vaquette, S. Saifzadeh, A. Farag, D. W. Hutmacher, S. Ivanovski, J. Dent. Res. 2019, 98, 673.

R. Oshana, DSP Softw. Dev. Tech. Embed. Real-Time Syst., Elsevier, Oxford, UK 2006.