Photoacoustic viscoelasticity imaging for the detection of acute hepatitis: a feasibility study

Biomechanical assessments are essential for the understanding of physiological states and the characterization of certain tissue pathologies such as liver cirrhosis. In this work, we showed by the photoacoustic viscoelasticity (PAVE) imaging that obvious mechanical change was also observed in the development of the acute hepatitis owing to the hepatocyte enlargement and intracellular ﬂuid increment, indicating that the PAVE technique can be developed as a supplementary method for detecting acute hepatitis in future. The feasibility of the PAVE imaging is validated by a group of agar phantoms. Furthermore, acute hepatitis pathological animal models were established and imaged ex vivo and in situ by the PAVE technique to demonstrate its capability for the mechanical characterization of acute hepatitis, and the imaging results were consistent with pathological results. The feasibility study of detecting acute hepatitis by the PAVE technique proved that this method has potential to be developed as a clinical biomechanical imaging method to supplement current clinical strategy for liver disease detection.


INTRODUCTION
Hepatitis infection is an important public health problem worldwide because hepatitis is one of the major causes of chronic hepatitis, cirrhosis, and hepatocellular carcinoma (Alter et al. 1999;Darby et al. 1997;Detre et al. 1996;Gao et al. 2015;Seeff et al. 1992;Tong et al. 1995). It is reported that more than two billion people worldwide are estimated to have evidence of current or past infection with hepatitis (Gao et al. 2013;Vedio et al. 2013), and usually some of them are asymptomatic. Liver function test (LFT) and liver biopsy (LB) are routine examination to diagnosis hepatitis. However, there are still problems in LFT such as low sensitivity of the enzyme detection in excess alcohol users (Yano et al. 2001). Even though LB is still considered as the gold standard in the evaluation of liver disease, it is invasive, painful, and costly, which is not an ideal method for repeated assessment of liver disease progression such as acute hepatitis detection (Ferraioli et al. 2012). In the development of the acute hepatitis, hepatocyte enlargement and intracellular fluid increment usually lead to tissue edema, which potentially induces the changes of liver viscoelasticity. Therefore, the method that can detect the mechanical properties of acute hepatitis can be an alternative way to supplement current clinical strategy.
Photoacoustic (PA) imaging is a hybrid imaging modality that combines the advantages of high optical contrast and high ultrasonic resolution Ermilov et al. 2009;Han et al. 2013;Hu et al. 2009;Lee et al. 2018;Song et al. 2014;Wang and Hu 2012;Yang et al. 2007a, b;Zeng et al. 2004;Zhang et al. 2014;Zhong et al. 2013). When excited by incident laser, tissues absorb photons and convert them into PA waves through thermal expansion (Gao et al. 2014;Liu and Yuan 2016;Yuan et al. 2012). As mechanical waves, the produced PA signal inherently carries mechanical information of the excited tissue (Gao et al. 2011(Gao et al. , 2017Hsieh et al. 2012;Singh and Jiang 2014;Yang et al. 2007a, b;Yuan et al. 2010), enabling PA imaging to be a potential strategy for tissue biomechanical characterization. Our group has proposed the PA viscoelasticity (PAVE) imaging for noninvasively characterizing tissue biomechanics through phase-resolved viscoelasticity ratio (VER) measurement (Gao et al. 2011;Shan et al. 2018), and has applied it to atherosclerosis characterization, tumor detection (Lv et al. 2018;Wang et al. 2004;Zhao et al. 2014Zhao et al. , 2016, and liver fibrosis detection (Wang et al. 2018). In this study, the feasibility of this method for acute hepatitis detection is provided. Animal liver models are established and imaged ex vivo and in situ, demonstrating the method to be a potential technique supplement to clinical acute hepatitis assessment. Figure 1A shows the principle of the laser-induced PA signal. When the tissue is excited by a sine-modulated laser beam with laser energy density of I ¼ I 0 sin xt ð Þ, the tissue absorbs incident photons and causes a temperature variation with a form of T ¼ T 0 sin xt ð Þ. The sine-modulated temperature field will induce thermal expansion which then induces sinusoidal PA wave generation with the same modulation frequency of the incident laser. In the PA signal production process, the cyclical heating induces thermal stress in the local region, and then the strain is generated. Because of the viscoelastic damping effect of biological tissues, there will be a phase lag between the incident laser and the produced PA signal, as shown in Fig. 1B. According to the Kelvin-Voigt model, the phase delay corresponds to the tissues viscosity-elasticity ratio can be written as (Gao et al. 2011;Shi et al. 2017):

PRINCIPLE OF THE PAVE IMAGING
where d is the PA phase delay, x is the laser modulation frequency, E is the Young's modulus, and g is the tissue viscosity, respectively. In this work, we validated the feasibility of the PAVE technique for acute hepatitis detection. The pathological and mechanical changes of acute hepatitis are shown in Fig. 1C. When the liver tissue is inflamed, inflammatory cells grow and infiltrate normal liver cells. This induces liver cell enlargement and increases intracellular fluid and edema, and the biomechanical properties such as the viscoelasticity ratio of the liver tissue may suffer obvious changes. Therefore, the PAVE imaging can be used for the detection of acute hepatitis.

THE CONSTRUCTION OF THE PAVE IMAGING SYSTEM
The systematic diagram of the PAVE imaging is shown in Fig. 2A. A high repetition frequency quasi-continuous laser (DS20HE-1064D/R, Photonics Industries) with a wavelength of 1064 nm and repetition frequency of 500 kHz is used as the excitation source. After laser collimation, the laser beam is focused by a microscope objective (NA = 0.1) to illuminate the tissue sample which is fixed on a two-dimensional stage and immersed in a water tank. The produced PA signal is then detected by an ultrasound transducer with a flat spectrum from 200 kHz to 15 MHz (HPM05/3, Precision Acoustics Ltd.). The detected signal is then transferred to an amplifier (ZFL-500, Minicircuits) for signal amplification and then acquired by a lock-in amplifier (SR830, Stanford Research Systems) to obtain the viscoelasticity-induced PA phase delay. By point-to-point scanning, the PAVE image then can be obtained through a MATLAB program. The twodimensional stage is controlled by the computer with a custom program written by LabVIEW (National Instruments, USA). The photo of the PAVE imaging set-up is shown in Fig. 2B. In order to prove that the proposed PAVE imaging inherits the high-resolution imaging capability of conventional PA techniques, two carbon tubes were imaged by the PAVE imaging system, and the result is shown in Fig. 2C. The resolution experiment as shown in Fig. 2D indicates that the lateral resolution of the PAVE imaging system is about 6.4 lm, which is capable of providing micro-mechanical information of target sample with high resolution.

THE CAPABILITY OF THE PAVE IMAGING SYSTEM
To demonstrate the capability of the PAVE imaging in biomechanical characterization, tissue-mimic agar phantoms were prepared and tested. Four agar phantoms with different agar concentrations and India ink concentrations with 1% agar and 5% ink, 2% agar and 5% ink, 3% agar and 2% ink, 4% agar and 2% ink are shown in Fig. 3A. It is reported that the mechanical properties of agar phantoms are highly correlated with the agar concentrations (Hall et al. 1997;Singh and Jiang 2014). The conventional PA imaging of the four agar phantoms shown in Fig. 3B reflects sample's absorption that corresponds to the ink concentration. As shown in Fig. 3C Fig. 3E) were imaged by the PAVE imaging system, and the result is shown in Fig. 3F. Owing to the fact that more fibrin and less fat content are existed in liver tissue, the phase delay of the liver tissue is relatively small compared with that of the fat tissue. Figure 3G shows the phase delay along the dotted line in Fig. 3F. These phantom and ex vivo tissue imaging results demonstrate that the PAVE imaging is an effective method for biomechanical characterization. The modeling process of acute hepatitis Sprague Dawley (SD) rat models with different pathological states is shown in Fig. 4A. 2 mL (10 mL/kg) carbon tetrachloride (CCl 4 ) olive oil solution with 0.1% concentration was used and intraperitoneally injected to modeling the acute hepatitis (Campo et al. 2008). The mechanism of hepatic injury-induced acute hepatitis by CCl 4 injection was reported mainly owing to the free radical during its catabolism (LeSage et al. 1999). The generation of trichloromethyl radicals and trichloromethyl peroxy radicals due to CCl 4 injection can attack the phospholipid molecules on the cell membrane of the liver, which induces cell membrane injury. Meanwhile, the radicals can also combine with membrane lipid and protein molecular to destruct the protein metabolism. By this way, the injection of the mixture of CCl 4 and olive oil will finally induce the death of the liver cells and acute hepatitis. In our experiments, six rats were injected with the mixture of CCl 4 and olive oil, and were divided into three groups, and two of them were dissected soon after CCl 4 injection to provide as liver tissue control group. The other two groups of rat The PAVE imaging for the detection of acute hepatitis METHOD were dissected after 24 and 48 h after CCl 4 injection to provide acute hepatitis tissue with different pathological states (Campo et al. 2008). After 24-h injection of CCl 4 solution, the concentration of alanine aminotransferase (ALT) in liver cells reached highest value, indicating that the liver suffers the most acute hepatitis. The photos of liver tissue with acute hepatitis are shown in Fig. 4B. Results in Fig. 4C show that the viscoelastic phase delay significantly increases compared with normal liver. After 48-h injection, the concentration of ALT decreased due to immune self-healing, and the liver viscosity began to reduce. The viscosity images agree well with the histological results shown in Fig. 4D. Since the acute hepatitis model is a rapid-form process, the fibrin has been barely deposited in the liver, and thus the liver elasticity has little obvious changes, as demonstrated by the histological results in Fig. 4E. The statistical results of the liver phase delay and the variation of ALT in Fig. 4F indicate that the PAVE imaging can be used as a potential method for assessment of acute hepatitis. Compared with existed clinical mechanical techniques, our method is capable of providing comprehensive tissue biomechanics with cellular resolution, and can be easily implemented into conventional PA microscopy, endoscopy, and laparoscope systems. To further demonstrate its capability for liver biomechanical characterization, in situ liver tissues were imaged by the PAVE imaging. As shown in Fig. 5A and B, mouse models with normal and 24 h acute hepatitis livers were executed to expose the liver tissue for PAVE imaging. The obtained PAVE images corresponding to the dotted line frame in Fig. 5A and B are shown in Fig. 5C and D. It is shown in the PAVE images that the PA phase delay of the acute hepatitis liver is much larger than that of the normal liver, which agrees well with the statistical results in Fig. 5G and H. The corresponding ultrasound elastography (USE) in Fig. 5E and F shows that the elasticity of the liver has little changes during the acute hepatitis. This is because the acute hepatitis is a rapid process of mold formation, and thus there is no obvious deposition of fibrous components in the liver tissue. The comparison of the PA phase delay and the USE demonstrates that the PAVE imaging is an effective strategy for the assessment of acute hepatitis.

DISCUSSION
In this work, by using the phase-sensitive image contrast, the PAVE imaging is able to differentiate subtle changes in the biomechanical characteristics of the acute hepatitis, which can allow effective detection of acute hepatitis and help to improve the detection accuracy in clinic. Nevertheless, there are still some limitations existed in the PAVE imaging. First, as we used a lock-in amplifier to obtain the PA phase delay, the imaging speed was restricted by the time constant (30 ms) of the lock-in amplifier. By reducing the time constant, higher imaging speed then can be obtained with a sacrifice of the signal-to-noise ratio. Second, due to distance-dependence phase deviation for the PA phase detection, ultrasonic localization will be necessary to calibrate the phase errors resulting from the sample surface roughness, which is essential to realize in vivo experiments. In this work, a 2.5-mm-sized broadband transducer with a flat spectrum from 200 kHz to 15 MHz was used to detect the PA signal. By using a narrowband transducer, the signal-to-noise ratio (SNR) in PAVE imaging can be further improved.

CONCLUSION
In summary, we present a method for the detection of the acute hepatitis by the PAVE imaging. The feasibility of the PAVE imaging was validated by a group of agar phantoms. Acute hepatitis animal models were established and imaged ex vivo to demonstrate the capability of the PAVE imaging for effective detection of the acute hepatitis. In situ animal models were imaged to further demonstrate the capacity of the proposed method in future biomedical applications. Overall, the experiment results show that the PAVE imaging has potential to be developed as a clinical biomechanical imaging method to supplement current clinical strategy for liver disease detection. 2014B020215003, 2014A020215031), the Postdoctoral Innovative Talent Support Program of China (BX201700084), and the Distinguished Young Teacher Project in Higher Education of Guangdong, China (YQ2015049). We thank the Third Affiliated Hospital of Sun Yat-Sen University for providing the results of shear wave elastography.

Compliance with Ethical Standards
Conflict of interest Qian Wang and Yujiao Shi declare that they have no conflict of interest.
Human and animal rights and informed consent All institutional and national guidelines for the care and use of laboratory animals were followed.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/ licenses/by/4.0/.