Acoustic Force Elastography Microscopy

Abstract

Objective: Hydrogel scaffolds have attracted attention to develop cellular therapy and tissue engineering platforms for regenerative medicine applications. Among factors, local mechanical properties of scaffolds drive the functionalities of cell niche. Dynamic mechanical analysis (DMA), the standard method to characterize mechanical properties of hydrogels, restricts development in tissue engineering because the measurement provides a single elasticity value for the sample, requires direct contact, and represents a destructive evaluation preventing longitudinal studies on the same sample. We propose a novel technique, acoustic force elastography microscopy (AFEM), to evaluate elastic properties of tissue engineering scaffolds.

Results: AFEM can resolve localized and two-dimensional (2D) elastic properties of both transparent and opaque materials with advantages of being non-contact and non-destructive. Gelatin hydrogels, neat synthetic oligo[poly(ethylene glycol)fumarate] (OPF) scaffolds, OPF hydroxyapatite nanocomposite scaffolds and ex vivo biological tissue were examined with AFEM to evaluate the elastic modulus. These measurements of Young's modulus range from approximately 2 kPa to over 100 kPa were evaluated and are in good agreement with finite element simulations, surface wave measurements, and DMA tests.

Conclusion: The AFEM can resolve localized and 2D elastic properties of hydrogels, scaffolds and thin biological tissues. These materials can either be transparent or non-transparent and their evaluation can be done in a non-contact and non-destructive manner, thereby facilitating longitudinal evaluation.

Significance: AFEM is a promising technique to quantify elastic properties of scaffolds for tissue engineering and will be applied to provide new insights for exploring elastic changes of cell-laden scaffolds for tissue engineering and material science.

Fig. 1.

Principle of Acoustic Force Elastography Microscopy (AFEM). The LPSWs are generated by the acoustic radiation force, and the RLPSW traveling through the sample thickness h is detected at the surface by OCT. GM: galvanometer mirror, RF: radio frequency, LPSW: longitudinally polarized shear waves, RLPSW: reflected longitudinally polarized shear waves, OCT: optical coherence tomography, TOF: time-of-flight.

Fig. 2.

System structure of acoustic force elastography microscopy (AFEM). The three function generators are utilized to provide trigger and driving signals for synchronizing the AFEM scan (Function Generator 1), acoustic excitation (Function Generator 2) and recording signals with the OCT system (Function Generator 3). The details of red dashed rectangle are illustrated in Fig. 1 to explain the behaviors of LPSWs and RLPSWs. The LPSWs are generated by the acoustic radiation force and travel through a thin scaffold with thickness h. The RLPSWs are detected at the surface by OCT. GM: galvanometer mirror, RF: radiofrequency.

Fig. 3.

An illustration of the AFEM scan pattern. The elastography resolution in the lateral and elevational direction on a scaffold with X by Y mm scan range represents Δk and Δs, respectively. p: an individual elastography measurement position, i: the number of M-scans, j: the number of B-scans, z: depth, k: total scan numbers of lateral elastography measurement positions, s: total scan numbers of elevational elastography measurement positions, Δk: elastography resolution in the lateral direction, Δs: elastography resolution in the elevational elastography direction, X: sample width, Y: sample length.

Fig. 4.

Numerical simulations compared with the AFEM results. Panels (a) and (b) illustrate the LPSW and the RLPSW, detected by AFEM, traveling vertically inside the 20% v/v tissue-mimicking gelatin phantom with 4.5 mm thickness, respectively. The video of the wave propagations can be found in the Movie S1. The LPSW and the RLPSW observed in the numerical simulation are presented in (c) and (d) in a material with the same mechanical properties of the 20% v/v gelatin phantom, respectively. The video of the numerical simulation is contained in Movie S2. Panels (e) and (f) show the spatiotemporal map from the experiment and the numerical simulation, respectively. The profiles selected at the excitation position indicated that the timing of RLPSW between the experiment and the simulation is highly consistent, shown in (g). Panel (h) illustrates the phase velocities calculated using the surface waves occurred on the 20% v/v gelatin phantom and the numerical simulation.

Fig. 5.

AFEM demonstrated on samples with fixed thickness. The RLPSWs were detected in (a) 8% v/v, (b) 13% v/v and (c) 18% v/v tissue-mimicking gelatin phantom with a constant thickness of 6 mm. The profiles selected at the excitation position indicated that the detected timing of RLPSWs is inversely proportional to the stiffness of the material, shown in (d).

Fig. 6.

AFEM demonstrated on samples with fixed stiffness. The RLPSWs were detected in tissue-mimicking gelatin phantoms with varying thicknesses of (a) 4.0 mm, (b) 3.2 mm and (c) 2.5 mm while maintaining a constant stiffness with an 8% v/v gelatin concentration. The profiles selected at the excitation position indicated that the detected timing of RLPSWs is proportional to the thickness of the material, shown in (d).

Fig. 7.

The elasticity measurement of porcine kidney tissue using AFEM. Panel (a) shows the porcine kidney cortex with the approximate 1.53 mm thickness as measured with OCT. The spatiotemporal map reveals the RLPSW and the surface wave propagating in the kidney tissue, presented in (b). The profiles from ten acquisitions at the same location indicate that the detected arrival time of RLPSWs is highly reproducible, shown in (c).

Fig. 8.

The 2D elasticity measurement of the neat transparent OPF scaffolds and OPF composite scaffolds with 10% w/w HA by using AFEM. Panel (a) shows a neat transparent OPF scaffold with M_n of 10,000 g/mol. (b) The thickness of the scaffold was imaged with OCT. Panels (c) and (d) show the RLPSW measured by AFEM and the numerical simulation result at a single position of the scaffold, respectively. (e) illustrates the reflected times of the RLPSWs with a total of 25 different locations (5 × 5 measurements in lateral and elevational direction with 2.54 mm elastography spacing) using AFEM. The scan range of the 2D measurement was approximately 12.7 mm × 12.7 mm, indicated as the rectangle with the dark red color in (a). The t_a over the 2D region was 0.31 ± 0.003 ms. (f) displays the distribution of the thickness of the scaffold over the 2D region and the h was 0.83 ± 0.006 mm. The 2D C_AFEM map was illustrated in (g) and C_AFEM was 5.35 ± 0.07 m/s. In panel (h) the E_AFEM of the scaffold was obtained and compared with the DMA results shown in (i). (j) An OPF composite scaffold with 10% w/w HA. (k) The RLPSW of the OPF composite scaffold at a single position was detected with AFEM. (l) The E_AFEM from three measurements at the three different positions were obtained and compared the AFEM results with the DMA tests.

Fig. 9.

Evaluation of the elastography resolution of AFEM in a gelatin hydrogel with high mechanical contrast. Two planes with approximately 2 mm distance were sliced using a tissue slicer blade to make a thin gelatin hydrogel sample and boundary was indicated with white dash line, presented in (a). A total of 21 measurements with the step of 0.635 mm were performed and the 11^th measurement was exactly at the boundary. The wave velocity profile was fit to a sigmoid function as an edge-spread function (b) for the calculation of the elastography spatial resolution of AFEM based on full-width half-maximum (FWHM) of the differentiated edge-spread function and the elastography spatial resolution was approximately evaluated as 720 μm.

Fig. 10.

Evaluation of the potential effects of the ARF causing vibration of the Petri dish bottom to influence RLPSW results in AFEM. Panel (a) shows a customized bottom with Mylar film. Panel (b) shows the RLPSW and the surface wave traveling in the gelatin hydrogel placed in the regular Petri dish and (c) placed in the Petri dish bottom replaced by a Mylar film. Panel (d) showed the phase velocity of the surface wave propagations on both phantoms.