Open Access
Add to BibSonomy
Abstract
Neonatal brain hemorrhage (NBH) is the most common neurological disorder in neonates and its clinical interventions are very limited. Understanding the pathology of NBH by non-invasive in-vivo characterization of standardized animal models is essential for developing potential treatments. Currently, there is no suitable tool to provide non-invasive, non-ionizing dynamic imaging of neonatal mouse models with high resolution, high contrast, and deep imaging depth. In this study, we implemented a fast 3D photoacoustic tomography (PAT) system suitable for imaging neonatal mouse brains with good image quality and demonstrated its feasibility in non-invasive monitoring of the dynamic process of NBH in the whole neonatal mouse brain. The results present a high resolution and sensitivity for NBH detection. Both morphological and hemodynamic changes of the hematoma were accurately obtained. Our results demonstrated the potential of PAT as a powerful tool for the preclinical study of neonatal brain hemorrhage.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Neonatal Brain hemorrhage (NBH) is a major consequence of preterm birth and the most common neurological disorder in neonates. NBH has become an important socio-economic problem due to the rise in preterm birth rates [1]. Infants who suffered NBH often develop hydrocephalus and consequential neurological deficits, such as seizure, cerebral palsy, cognitive or learning deficits [2–4]. However, clinical interventions of NBH are still very limited [5,6], therefore, understanding the mechanisms of NBH pathology such as its progression in brain tissue and the consequential morphological and hemodynamic alterations, by characterizing standardized animal models is essential for developing potential treatments.
The use of NBH animal models provides great value in physiopathology study and treatment evaluation, due to the nature of the injury progression study, the complexity of human subjects, and the requirement of controlled experimental conditions [1, 7, 8]. Among all the animal models, rodents have better neurodevelopmental similarities to preterm human infants, and neonatal mice and rats have been used as standardized animal models for human neonatal germinal matrix hemorrhage studies [9]. Previous studies have shown the differences between neonatal and adult mouse brain hemorrhage models [10]. Therefore, the neonatal mouse is required for studying the NBH due to the age-dependent brain responses in the germinal matrix hemorrhage (GMH) and intraventricular hemorrhage (IVH) which are typical NBH types.
Imaging the dynamic process of neonatal mouse brain hemorrhage is challenging due to the requirements of high spatial and temporal resolution, high imaging contrast of blood vessels and hematoma, and imaging depth deep enough to cover the whole brain. Currently, characterizing NBH models primarily relies on traditional imaging techniques, such as magnetic resonance imaging (MRI), computed tomography (CT), and ultrasonography (US). However, there are some limitations in these techniques. MRI is very costly due to the requirement of a high magnetic field for high-resolution imaging of the neonatal mouse brain [11]. In addition, the low temporal resolution makes it unsuitable for dynamic imaging. CT employs ionizing radiation which limits its application in neonates and infants. The US imaging can offer real-time imaging ability, however, its low acoustic specificity leads to low contrast, and researches show that contrast enhance diagnostic ultrasound has the risk to cause microvascular injury and hemorrhage [12]. Optical imaging offers high resolution and optical contrast, however, the imaging depth is very limited which cannot be applied to whole-brain imaging [13].
To address the above challenges, a non-invasive imaging technique that offers a high spatial and temporal resolution, high contrast, and deep imaging depth is needed for NBH study. Photoacoustic tomography (PAT) which combines high acoustic resolution and high optical contrast, provides an ideal solution. As hemoglobin is the endogenous contrast agent which provides the strongest PA signal in the near-infrared region [14,15], PAT is an ideal imaging tool for hemorrhage detection. Besides, PAT provides functional imaging ability, which assesses critical pathophysiological parameters such as oxygen saturation ($s{O_2}$), hemoglobin concentrations (HbT), cerebral blood flow (CBF) [16–21]. These important parameters can be used to evaluate the grades of hemorrhage and monitor brain recovery. In addition, PAT can offer real-time dynamic imaging to monitor the progression of the hemorrhage model, which will deliver valuable insights for understanding the physiopathology of neonatal brain hemorrhage. Although there are previous studies in an adult mouse IVH model to observe the hematoma in a single brain layer. The single transducer PAT systems used in those studies showed a very low spatial and temporal resolution [22,23], which is not sufficient for imaging neonatal mouse brains. By now, there is no study using PAT to monitor the dynamic progression of the whole brain hemorrhage in a neonatal mouse model.
There are some challenges for imaging neonatal mouse models. Previous studies have shown that PAT can be applied for dynamic imaging of adult mice/rat brains [22,24–26]. The results show the major blood vessels and the dynamic change of the brain. However, in most results, the microvasculature or smaller blood vessels are usually missing. Compared to adult mice, the blood vessels are way smaller for neonatal mice (postnatal day zero). Therefore, better image quality is required for PAT to be applied in the study of NBH using neonatal mouse models. Although optical resolution photoacoustic microscopy (OR-PAM) has a very high resolution to image microvasculature, the imaging depth is very limited (∼1 mm) [27], which is not suitable for imaging the whole neonatal brain. Thus, developing a PAT system that is suitable to image neonatal mouse models is necessary for the study of NBH and other congenital brain diseases, in which adult mouse models cannot be used.
Another challenge in this application is acoustic coupling. Unlike adult mice, which have hard and thick skulls to protect the brain, the neonatal mouse skull is very thin and fragile. Applying forces to it would alter the blood perfusion or even damage the brain. Thus, the coupling methods in the previous PAT systems for brain imaging are not feasible in this application [25,28]. Therefore, A 3D PAT system that is suitable for NBH studies in neonatal mouse brains is demanding.
In this study, we implemented a fast 3D PAT system for imaging neonatal mouse models and demonstrated its feasibility to monitor and evaluate the progression of NBH in the whole neonatal mouse brain. The dynamic changes of the lesion in the whole brain were assessed. The hemoglobin evolution process and oxygen saturation change of the hematoma were evaluated quantitatively. To our best knowledge, this study reports the first noninvasive assessment of the dynamic process of NBH in the whole neonatal brain using PAT, which demonstrated its potential as a powerful tool for studying the physiopathology of neonatal brain hemorrhage.
2. Materials and methods
2.1 Neonatal GMH/IVH model
CD-1 mouse pups of postnatal day zero (P0) were anesthetized with 3% isoflurane in mixed air and oxygen. The pups are placed on a stereotaxic frame for collagenase injection. 0.3 units collagenase VII-S (Sigma, St Louis, MO) was injected into the neonatal mouse brain through a 33-gauge Hamilton syringe (0.210 mm outer diameter). The needle remained in place for 5 minutes after injection to prevent “back-leakage”. Mice were injected with collagenase in the periventricular germinal matrix area (1 mm lateral, 0.6 mm behind the eye, and 3 mm below the scalp surface).
2.2 Imaging system
The fast 3D photoacoustic imaging system in this study is shown schematically in Fig. 1. In this system, a portable fast-tuning OPO laser (Phocus Mobile, Opoteck, Carlsbad, California, wavelength 690-950 nm, frequency 20 Hz) was used as the excitation source. The laser beam was coupled into a high energy fiber bundle optimized for near-infrared tuning ranges. The built-in energy meter monitors the OPO pulse energy in real-time and provides feedback for harmonics auto-optimization and logs pulse energy for data normalization. The delivered laser energy density to the animal was within 22mJ/$c{m^2}$ of ANSI safety limit for 720 nm wavelength (the maximum permissible exposure is calculated based on the guidance in the American National Standard for Safe Use of Lasers). A custom-built hyperbolic ultrasound array probe (Japan Probe Co Ltd, Japan, 256 elements, array radius: 65 mm; central frequency: 4.4 MHz; bandwidth 113%) was used for tomographic PA imaging. The schematic of the cylindrically focused ultrasound transducer array (L = 20 mm, W = 0.7 mm, Ra = 65 mm, Rse = 58 mm) is shown in Fig. 1(B). Data were amplified and recorded with a custom-made 256-channel amplifier/DAQ system (PhotoSound Technologies Inc., Houston, Texas), and transferred in realtime to a computer via a USB 3.0 interface. The water in the tank was kept around 37 °C and was continuously monitored through a thermometer (National Instruments, USB-TC01) with a J-type thermocouple. The logged temperature data was used for PAT image reconstruction using a sound velocity calibrated algorithm to eliminate the artifacts induced by temperature changes. For each laser pulse, a 2D tomographic image can be obtained, and 3D imaging was achieved by scanning the probe vertically. It takes about 4s to scan the whole neonatal brain with a 5 mm depth and step size of 0.5 mm. The temporal resolution of tomographic imaging is limited by the laser repetition rate (20 Hz), therefore the frame rate is 20fps. The quantified spatial resolution is 105$\mu m$.
Fig. 1. Fast 3D photoacoustic system for in vivo neonatal mouse brain imaging. A) system schematics; B) transducer array design; AP: Amplifier, BM: Breathing mask, DAQ: Data acquisition, HT: Heater, LM: Linear translation step motor, MH: Mouse holder, MS: Moving stage, RM: Rotator step motor, TM: Thermoelectric thermometer, UT: Ultrasonic transducer array; Ra: radius of the array; Rse: radius of a single element; L: element length; P: pitch of each element; W: element width.
To realize PAT imaging of neonatal mouse brains, we optimized the system to achieve better image quality. The transducer array was designed to optimize the sensitivity, aperture angle, tomographic resolution, and FOV. The transducers were made of a piezoelectric material that has better receiving characteristics compared to commercially used materials, and thus, it has better sensitivity and broader bandwidth. The signal-to-noise ratio (SNR) was improved through system design of reducing the electrical and acoustic noises to realize single-pulse imaging without sacrificing image quality.
Acoustic coupling for in-vivo neonatal mouse brain imaging is also challenging. Since the neonatal mouse is very small, to image the mouse brain, the head needs to be immersed in the coupling media which makes it challenging to deliver air for breathing. To address this problem, we designed and fabricated a waterproof sealed mask for neonatal mice, and used it for air and anesthetizer delivery. Mice were immersed in water for acoustic coupling in PAT imaging. This design also allows us to achieve better PAT image quality as it avoids the use of plastic membrane and ultrasound gel which interfere with light and ultrasound transmission in previous PAT brain imaging setups.
2.3 Photoacoustic imaging of neonatal brain hemorrhage progression
The in-vivo whole neonatal mouse brains were imaged noninvasively for two hours after the injection of collagenase. Tomographic images were acquired at every imaging plane in real-time, and the whole mouse brain was imaged by scanning the probe in the vertical direction. The mice were anesthetized with 3% isoflurane in mixed air and oxygen during the whole imaging process. The animal experiments in this study were conducted in accordance with the research ethics board at the University of South Florida.
2.4 Image reconstruction and multispectral PAT of oxygen saturation evaluation
Images were reconstructed with the PAT data using a temperature-calibrated delay and sum algorithm. For blood vessel and structure visualization, a bandpass filter was applied to enhance the contrast of the small structures. For quantitative evaluation, the original signals were used, to ensure accuracy. The evolution of pathophysiological parameters including oxygen saturation ($s{O_2}$) were imaged using multispectral PAT. Wavelengths of 780 nm and 850 nm which were fast-tuned at each laser pulse were employed in this study. The results were unmixed using least-squares spectral fitting to obtain the concentration of HbO and HbR. Oxygen saturation ($s{O_2}$) was calculated as $s{O_2} = {C_{HbO}}({x,y} )/({{C_{HbO}}({x,y} )+ {C_{HbR}}({x,y} )} )$ [29]. The least-square method can be described as the following:
By solving the linear equations of multiple wavelengths, ${C_{HbR}}({x,y} )$ and ${C_{HbO}}({x,y} )$ can be calculated:
2.5 Assessment of hematoma area
The lesion size during the injury process was assessed quantitatively using the PAT data. The lesion area was identified automatically. Due to the irregular shape of the hematoma, MATLAB was used to evaluate the lesion area by subtracting the data acquired at different time points during the hemorrhage from the data of the baseline. The light fluence was normalized. The images were registered before subtraction to eliminate the motion-induced error. Then the size of the lesion area was calculated by taking into account the pixels that have an intensity larger than the half maximum. The ratio of the lesion area to the brain area was calculated by diving the pixel number of the lesion by the pixel number of the entire brain area.
2.6 Statistical analysis
The relative change of lesion size, and oxy- and deoxyhemoglobin change over time were calculated and statistical significance was analyzed using one-way ANOVA (p < 0.05 was considered statistically significant).
3. Results
3.1 In vivo neonatal mouse brain imaging
Tomographic images of the whole neonatal mouse brain were acquired to validate the ability of our fast 3D PAT system. Representative tomographic images are shown in Fig. 2. The cerebral blood vessels and structures of different brain slices can be clearly seen in the PAT images, which allows us to monitor and analyze the NBH process.
3.2 In vivo monitoring of the NBH process
The representative tomographic image of the same imaging plane at different time points during the hemorrhage progression is shown in Fig. 3. The lesion can be clearly seen in the PAT, and is in good agreement with its corresponding histological section. The yellow arrow indicates the hemorrhage lesion, and the red arrow indicates the needle hole. The low optical absorption region (red dashed line) indicates the reduction of cerebral blood flow in this region, which tends to become local ischemia penumbra.
Fig. 3. Representative photoacoustic images of hematoma lesion progression with comparison to histology. The red arrow indicates the needle injection site, the yellow arrow indicates the hematoma, and the red dashed lines indicate the local ischemia; the scale bar is for PAT images.
3.3 Quantitative assessment of hematoma lesion development
Figure 4(A). shows the representative tomographic images of the lesion size progression. The lesion was identified automatically and was quantitatively analyzed. From these images, we can see that the hemorrhage started in the right periventricular region and gradually developed into the lateral and fourth ventricles. The dynamic change of lesion to brain size ratio in the first 2 hours of the hemorrhage is shown in Fig. 4(B). It shows that the hematoma region increased with time.
Fig. 4. Quantitative assessment of hematoma lesion size progression. A) representative tomographic images of the lesion size progression. B) the dynamic change of lesion-to-brain size ratio over time.
3.4 Monitoring the whole brain hematoma progression and lesion oxygen saturation change
Figure 5 presents the 3D progression of NBH in the whole neonatal mouse Brain. Figures 5(A) and 5(C) are representative 3D images at two different time points, and Fig. 5(B) and 5(D) are corresponding representative tomographic images at different depth of the whole brain. The progression of the lesion at different depths of the brain can be clearly seen in the PAT images.
The data of the 3D lesion-to-brain size ratio in the whole neonatal mouse brain for three mice are shown in Fig. 6(A). The data is presented as mean${\pm} $SD. The statistical significance of the change in lesion-to-brain size ratio was observed in all three mice (P < 0.05). To further analyze the process of NBH. We recovered the dynamic change of oxygen saturation in the hematoma area. We observed a decrease in oxygen saturation in the lesion. The percentage variation (SEM) of $s{O_2}$ overtime for this study (n = 3) is shown in Fig. 6(B). The $s{O_2}$ in lesion decreased by 4.978${\pm} $1.135%. There is no statistical significance in $s{O_2}$ variation in the first 2 hours of NBH. However, we observed a larger decrease in $s{O_2}$ 48 hours post NBH. The $s{O_2}$ decreased by 25.741${\; } \pm $ 11.722%. These results indicate that a significant decrease in oxygen saturation does not occur in the first few hours of NBH, and the change in oxygen saturation at different stages of NBH varies.
Fig. 5. 3D progression of NBH in the whole neonatal mouse Brain. A) and C) representative 3D images at different time points during hemorrhage. B) and D) corresponding representative tomographic images of different brain sections.
Fig. 6. Quantitative data of change in lesion-to-brain size ratio and the percentage variation of oxygen saturation over time (n = 3). A) dynamic change of lesion/brain size ratio in the whole brain over time. B) percentage variation of $s{O_2}$ overtime.
4. Discussion
In this study, we investigated the progression of neonatal brain hemorrhage using the collagenase-induced NBH model. The injection is in the right germinal matrix region. The needle hole can be clearly seen in the PAT image. The bleeding started from the germinal matrix and right lateral ventricle, then gradually progressed into the left lateral ventricle, followed by the fourth ventricle. The progression of the NBH in the horizontal tomographic brain section and the whole mouse brain are both evaluated, demonstrating the ability to monitor and evaluate the process of hematoma growth. The results showed a GMH with intraventricular extension and ventricular enlargement, which indicated a grade IV (severe) GMH as defined by clinical imaging studies in premature infants [30]. According to previous studies, the extent of bleeding is the essential factor that associates the most with morbidity and mortality [1,31]. Grade I to II hemorrhage related to developmental disabilities and grade III to IV related to long-term complications like mental retardation and cerebral palsy [1]. About 85% of the survivors of NBH developed major cognitive dysfunction [32,33]. Therefore, monitoring and quantitatively evaluating the process of bleeding progression and severity will give us more insights into the development of NBH and its complications.
The dynamic change of oxygen saturation in the hematoma area over time was investigated in this study. The results showed a small decrease in $s{O_2}$ during the first 2 hours of NBH. And a larger decrease after 48 hours post-hemorrhage. Clinically, IVH/GMH is induced by the rupture of tiny arteries, resulting in hematoma formation or blood clots in the lesion core, and periphery tissue distortion. We observed that the change of $s{O_2}$ varies at different stages of hemorrhage. At the beginning of the bleeding, HbO consists of a large portion of the hemoglobin, and with the progression of hematoma, the concentration of HbR increases, resulting in decreased oxygen saturation. IVH/GMH can induce hypoxia-ischemia, in addition, due to the lack of regional autoregulation in cerebral vasculature and insufficient vascularization in the periventricular region, the hypoxia-induced injury can be intensified. In the PAT images, the local ischemia in the periphery region of the hematoma can be clearly observed. The low optical absorption in the surrounding tissue indicates a reduction in cerebral blood flow during the acute phase of NBH. The hematoma can induce mechanical destruction like tearing of blood vessels in GM, and CBF may be compromised locally resulting in ischemic damage [34]. Oxygen metabolic and blood supply is essential for neurological development. Monitoring the alteration of these factors is important for understanding the mechanism of neurodevelopmental delay and deficits.
The neonatal brain hemorrhage model we investigated is analogous to clinical studies in premature human infants. In this model, the blood vessels in the GM region are ruptured and the ependymal is broken, leading to blood filling the ventricles. Compared to the model using injected blood, this model exhibits a progressive and spontaneous progression of focal bleeding, rebleeding, and blood vessel rupture, which are comparable to GMH/IVH of human premature infants [34,35].
In addition to hematoma progression and oxygen metabolism, brain function alteration is a very important complication of NBH. Our future study will investigate the post-hemorrhage brain functional connectivity alteration in the developing brain using PAT, and its correlation to the alteration of oxygen saturation, hematoma, and local ischemia.
5. Conclusion
In summary, we presented a fast 3D PAT system that is suitable for imaging neonatal mouse models and demonstrated its feasibility to non-invasively monitor and quantitatively evaluate the dynamic process of hemorrhage in the whole neonatal mouse brain. Morphological changes (such as location, shape, and size of the hematoma) during the hemorrhage process can be accurately obtained. Hemodynamic changes in the lesion and surrounding hemorrhagic area can be monitored. Our results present high resolution and contrast in the PAT imaging of hematoma and cerebral blood vessels in the neonatal mouse brain, as well as deep imaging depth to cover the whole brain, which demonstrates its potential as a powerful tool for the preclinical study of neonatal brain hemorrhage.
Funding
National Institutes of Health (5R01AA028200).
Disclosures
The authors declare no conflicts of interest. Portions of this work were presented at the Biophotonics Congress 2022, OS2D.2.
Data availability
Data may be obtained from the authors upon reasonable request.
References
1. T. Lekic, D. Klebe, R. Poblete, P. R Krafft, W. B Rolland, J. Tang, and J. J. C. M. C. H. Zhang, “Neonatal brain hemorrhage (NBH) of prematurity: translational mechanisms of the vascular-neural network,” Curr. Med. Chem. 22(10), 1214–1238 (2015). [CrossRef]
2. M. C. Allen and A. J. Capute, “Neonatal neurodevelopmental examination as a predictor of neuromotor outcome in premature infants,” Pediatrics 83(4), 498–506 (1989). [CrossRef]
3. P. Ballabh, A. Braun, and M. Nedergaard, “The blood–brain barrier: an overview: structure, regulation, and clinical implications,” Neurobiol. Dis. 16(1), 1–13 (2004). [CrossRef]
4. C. O. Chua, H. Chahboune, A. Braun, K. Dummula, C. E. Chua, J. Yu, Z. Ungvari, A. A. Sherbany, F. Hyder, and P. Ballabh, “Consequences of intraventricular hemorrhage in a rabbit pup model,” Stroke 40(10), 3369–3377 (2009). [CrossRef]
5. J. Yang, Q. Li, Z. Wang, C. Qi, X. Han, X. Lan, J. Wan, W. Wang, X. Zhao, and Z. Hou, “Multimodality MRI assessment of grey and white matter injury and blood-brain barrier disruption after intracerebral haemorrhage in mice,” Sci. Rep. 7(1), 1–12 (2017). [CrossRef]
6. G. Kenet, A. A. Kuperman, T. Strauss, and B. Brenner, “Neonatal IVH–mechanisms and management,” Thrombosis Res. 127, S120–S122 (2011). [CrossRef]
7. O. Adeoye, J. F. Clark, P. Khatri, K. R. Wagner, M. Zuccarello, and G. J. Pyne-Geithman, “Do current animal models of intracerebral hemorrhage mirror the human pathology?” Trans. Stroke Res. 2(1), 17–25 (2011). [CrossRef]
8. T. Lekic and J. H. Zhang, “Posterior circulation stroke and animal models,” Front Biosci 13(13), 1827–1844 (2008). [CrossRef]
9. T. Lekic, A. Manaenko, W. Rolland, P. R. Krafft, R. Peters, R. E. Hartman, O. Altay, J. Tang, and J. H. Zhang, “Rodent neonatal germinal matrix hemorrhage mimics the human brain injury, neurological consequences, and post-hemorrhagic hydrocephalus,” Exp. Neurol. 236(1), 69–78 (2012). [CrossRef]
10. M. Xue and M. R. Del Bigio, “Injections of blood, thrombin, and plasminogen more severely damage neonatal mouse brain than mature mouse brain,” Brain Pathol. 15(4), 273–280 (2006). [CrossRef]
11. D. Wu and J. Zhang, “In vivo mapping of macroscopic neuronal projections in the mouse hippocampus using high-resolution diffusion MRI,” NeuroImage 125, 84–93 (2016). [CrossRef]
12. X. Lu, C. Dou, M. L. Fabiilli, and D. L. Miller, “Capillary hemorrhage induced by contrast-enhanced diagnostic ultrasound in rat intestine,” Ultrasound Med. Biol. 45(8), 2133–2139 (2019). [CrossRef]
13. S. Aumann, S. Donner, J. Fischer, and F. J. Müller, “Optical coherence tomography (OCT): principle and technical realization,” in High Resolution Imaging in Microscopy and Ophthalmology: New Frontiers in Biomedical Optics, (Springer, 2019), 59–85.
14. Q. Chen, T. Jin, W. Qi, X. Mo, and L. J. Xi, “Label-free photoacoustic imaging of the cardio-cerebrovascular development in the embryonic zebrafish,” Biomed. Opt. Express 8(4), 2359–2367 (2017). [CrossRef]
15. H. Ma, Z. Cheng, Z. Wang, H. Qiu, T. Shen, D. Xing, Y. Gu, and S. J. Yang, “Quantitative and anatomical imaging of dermal angiopathy by noninvasive photoacoustic microscopic biopsy,” Biomed. Opt. Express 12(10), 6300–6316 (2021). [CrossRef]
16. T. Shan, Y. Zhao, S. Jiang, and H. J. Jiang, “In-vivo hemodynamic imaging of acute prenatal ethanol exposure in fetal brain by photoacoustic tomography,” J. Biophonotics 13(5), e201960161 (2020). [CrossRef]
17. T. Shan, J. Qi, M. Jiang, and H. J. A. O. Jiang, “GPU-based acceleration and mesh optimization of finite-element-method-based quantitative photoacoustic tomography: a step towards clinical applications,” Appl. Opt. 56(15), 4426–4432 (2017). [CrossRef]
18. L. Li, J. Yao, and L. V. Wang, “Photoacoustic tomography of neural systems,” in Neural Engineering (Springer, 2020), pp. 349–378.
19. B. Mc Larney, M. A. Hutter, O. Degtyaruk, X. L. Deán-Ben, and D. J. Razansky, “Monitoring of stimulus evoked murine somatosensory cortex hemodynamic activity with volumetric multi-spectral optoacoustic tomography,” Front Neurosci. 14, 536 (2020). [CrossRef]
20. J. Tang, L. Xi, J. Zhou, H. Huang, T. Zhang, P. R. Carney, and H. J. Jiang, “Noninvasive high-speed photoacoustic tomography of cerebral hemodynamics in awake-moving rats,” J. Cereb. Blood Flow Metab. 35(8), 1224–1232 (2015). [CrossRef]
21. M. Li, Y. Tang, and J. J. P. Yao, “Photoacoustic tomography of blood oxygenation: a mini review,” Photoacoustics 10, 65–73 (2018). [CrossRef]
22. D. Wu, J. Yang, G. Zhang, and H. J. Jiang, “Noninvasive in vivo monitoring of collagenase induced intracerebral hemorrhage by photoacoustic tomography,” Biomed. Opt. Express 8(4), 2276–2286 (2017). [CrossRef]
23. J. Yang, D. Wu, G. Zhang, Y. Zhao, M. Jiang, X. Yang, Q. Xu, and H. J. Jiang, “Intracerebral haemorrhage-induced injury progression assessed by cross-sectional photoacoustic tomography,” Biomed. Opt. Express 8(12), 5814–5824 (2017). [CrossRef]
24. J. Lv, S. Li, J. Zhang, F. Duan, Z. Wu, R. Chen, M. Chen, S. Huang, H. Ma, and L. J. Nie, “In vivo photoacoustic imaging dynamically monitors the structural and functional changes of ischemic stroke at a very early stage,” Theranostics 10(2), 816–828 (2020). [CrossRef]
25. L. Li, L. Zhu, C. Ma, L. Lin, J. Yao, L. Wang, K. Maslov, R. Zhang, W. Chen, and J. J. Shi, “Single-impulse panoramic photoacoustic computed tomography of small-animal whole-body dynamics at high spatiotemporal resolution,” Nat. Biomed. Eng. 1(5), 0071 (2017). [CrossRef]
26. W. Li, R. Chen, J. Lv, H. Wang, Y. Liu, Y. Peng, Z. Qian, G. Fu, and L. J. Nie, “In vivo photoacoustic imaging of brain injury and rehabilitation by high-efficient near-infrared dye labeled mesenchymal stem cells with enhanced brain barrier permeability,” Adv. Sci. 5(2), 1700277 (2018). [CrossRef]
27. S. Jeon, J. Kim, D. Lee, J. W. Baik, and C. J. Kim, “Review on practical photoacoustic microscopy,” Photoacoustics 15, 100141 (2019). [CrossRef]
28. M. Kuniyil Ajith Singh, W. Steenbergen, and S. Manohar, “Handheld probe-based dual mode ultrasound/photoacoustics for biomedical imaging,” in Frontiers in Biophotonics for Translational Medicine (Springer, 2016), pp. 209–247.
29. X. L. Deán-Ben and D. J. P. Razansky, “Functional optoacoustic human angiography with handheld video rate three dimensional scanner,” Photoacoustics 1(3-4), 68–73 (2013). [CrossRef]
30. L.-A. Papile, J. Burstein, R. Burstein, and H. Koffler, “Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm,” The Journal of Pediatrics 92(4), 529–534 (1978). [CrossRef]
31. A. Whitelaw, “Intraventricular haemorrhage and posthaemorrhagic hydrocephalus: pathogenesis, prevention and future interventions,” in Seminars in Neonatology (Elsevier, 2001), pp. 135–146.
32. B. R. Vohr, W. C. Allan, M. Westerveld, K. C. Schneider, K. H. Katz, R. W. Makuch, and L. R. Ment, “School-age outcomes of very low birth weight infants in the indomethacin intraventricular hemorrhage prevention trial,” Pediatrics 111(4), e340–e346 (2003). [CrossRef]
33. E. Roze, K. N. Van Braeckel, C. N. van der Veere, C. G. Maathuis, A. Martijn, and A. F. Bos, “Functional outcome at school age of preterm infants with periventricular hemorrhagic infarction,” Pediatrics 123(6), 1493–1500 (2009). [CrossRef]
34. M. Xue, J. Balasubramaniam, R. J. Buist, J. Peeling, and M. R. Del Bigio, “Periventricular/intraventricular hemorrhage in neonatal mouse cerebrum,” J. Neuropathol. Exp. Neurol. 62(11), 1154–1165 (2003). [CrossRef]
35. L. Siaudvytyte, A. Daveckaite, I. Januleviciene, A. Ragauskas, B. Siesky, and A. Harris, “Intracranial, intraocular and ocular perfusion pressures: differences between morning and afternoon measurements,” J. Model. Ophthalmol. 1(1), 37–50 (2016). [CrossRef]
