Open Access
Add to BibSonomy
Abstract
In this study, we in vivo examined injury progression after intracerebral haemorrhage (ICH) induced by collagenase in mice using cross-sectional photoacoustic tomography (csPAT). csPAT displayed high resolution with high sensitivity for ICH detection. The PAT images obtained showed high correlation with conventional histologic images. Quantitative analysis of the hematoma areas detected by csPAT showed high consistency with the neurologic deficit score (NDS). By utilizing the dual-wavelength method, the development of the hemoglobin area was monitored. Our results indicated that noninvasive csPAT can be used to track the dynamic progression of post-ICH, and to evaluate therapeutic interventions in preclinical ICH models.
© 2017 Optical Society of America
1. Introduction
Spontaneous intracerebral haemorrhage (ICH), having a mortality of 54.7%, remains a significant cause of morbidity and mortality throughout the world [1, 2]. Yet, interventions for ICH are still very limited [3]. So understanding the mechanisms of ICH pathology and repair, and identifying potential treatments have high priorities.
Preclinical ICH models have greatly improved our knowledge of the pathophysiology of ICH and have been a valuable tool for testing potential therapeutic strategies [4]. In particular, transgenic mouse models are helpful for addressing genetic influences on early brain injury and late brain repair after ICH [5]. Currently, imaging in ICH models typically relies on traditional methods like magnetic resonance imaging (MRI) and positron emission tomography (PET). Although these techniques provide deep penetration and multi-parameter information, they suffer from significant limitations. MRI requires a costly high magnetic field to achieve high resolution imaging at microscope scale and a long data acquisition time ranging from seconds to minutes, making imaging of dynamics difficult [6, 7]. PET on the other hand suffers from poor spatial resolution. To overcome all of the above limitations using one system, new imaging modalities are urgently needed.
Optical imaging of biological tissue employs non-carcinogenic electromagnetic waves to provide extraordinary structural, functional and molecular contrasts with either endogenous or exogenous agents [8–10]. Unfortunately, the application of conventional optical imaging technologies to small-animal deep brain imaging is impeded by the strong optical scattering, which prevents high-resolution imaging from beyond the optical diffusion limit of approximately 1–2 mm in depth [11]. Although diffusive optical imaging methods, such as fluorescence diffuse optical tomography [12], can provide centimeters of penetration, their spatial resolution is poor.
To date, photoacoustic tomography (PAT) is the only high resolution optical imaging modality that breaks the optical diffusion limit. It combines the advantages of optical contrast and acoustic detection, and holds great promise for small-animal imaging [13]. Previous works showed that PAT has great potential in various biological and medical applications such as vasculature visualization [14–17], arthritis diagnosis [20, 21], neuroimaging [23, 24].Our own previous work indicated that widely adopted horizontal plane PAT is a powerful tool to detect ICH morphology in small animals [17].
Unlike horizontal plane PAT, cross-sectional PAT (csPAT) is often used to acquire images in coronal plane in both mouse and rat models [16, 22]. It can be readily integrated into a multi-modal imaging system with ultrasound (US) and/or diffuse optical tomography (DOT) [19–21], offering more informed diagnosis of diseases. Furthermore, it is compatible with handheld operation, facilitating clinical translation with transducer array [23].
In this study, we implemented a csPAT system and evaluated its potential in noninvasively monitoring the injury processing in a mouse ICH model. In comparison with histologic examinations, we calculated the lesion area, analyzed the hemoglobin evolution process, and assessed the tissue loss in brain sections. This study, to the best of our knowledge, is the first noninvasive assessment of ICH injury progression using csPAT.
2. Methods
2.1 Experimental animals
The animal studies were conducted in accordance with the Research Ethics Board at the University of Electronic Science and Technology of China (UESTC). Twenty-one ICR mice (8 weeks old, 22-25g, 15 mice were used for csPAT image and 6 mice were used for functional assessment) were obtained from Chengdu Dossy experimental animal (Sichuan, China) and maintained in the animal facility. All efforts were made to minimize the number of animals used and their suffering. Before imaging, mice in each group were anesthetized by intraperitoneal injection of 1% pentobarbital sodium solution.
2.2 Intracerebral hemorrhage (ICH) induction
ICH model was induced by injecting collagenase into the right striatum (caudate putamen, CPu) of a mouse according to a well established method [3, 26]. A burr hole was drilled into the skull, and type VII-S collagenase (0.075 U in 0.5μl saline; Sigma, St. Louis, MO) was infused into the right striatum with a micro-infusion pump at a constant rate of 0.1 μl/min. The needle tip was placed for haematoma induction to invade the middle of the striatum without damaging the motor cortex: 0.2 mm anterior and 2.0 mm lateral from the bregma, 2.5 mm below the skull (Fig. 1(a)). Body temperature was maintained at 37 ± 0.5 °C during surgery. After surgery, the animals were returned to their cage.
Fig. 1 Modeling ICH in mice. (a) The simplified schematic of a coronal brain section 0.2 mm anterior of bregma illustrates the proposed location of collagenase injection. LV: lateral ventricle, CPu: caudate-putamen, cc: corpus callosum. (b) Representative photograph of a coronal brain section, obtained on day 3 after injection of collagenase.
2.3 Histological validation
Animals were sacrificed after experiment at days 1, 3, 7, and 28 after surgery. Their brains were excised and frozen after the experiments. To validate the lesion area, coronal brain slices were taken every 1 mm using a cryostat with two 10 μm sections being mounted on a metal slide [25]. No staining was applied here to get the frozen sections of a mouse brain.
2.4 Photoacoustic tomography system
The csPAT system used in this study is schematically shown in Fig. 2. In this system, a pulsed OPO laser (Surelite OPO, Continuum, CA, USA) pumped by a Q-switched Nd:YAG laser (Surelite I-20, Continuum, CA, USA) with a duration of ~4 ns and a repetition rate of 20Hz served as the excitation source. Through a beam splitter, 10% of the light energy was delivered to a photodiode (S120C, Thorlabs, USA) connected to a digital handheld optical power and energy meter console (PM100D, Thorlabs, USA) to monitor fluctuations of the laser energy. The remaining 90% of the laser energy was separated equally by a second beam splitter into two fiber bundles with line-shape output (NA: 0.2, 2mm x 6mm, Shangguang, Chengdu, China). The tips of the fiber bundles were positioned at the opposite sides of animal holder to maximize the illumination area. The laser energy density delivered to the animal was ~22 mJ/cm2, which is far below the ANSI safety limit of 100 mJ/cm2 in NIR region [27].
Fig. 2 Experimental configuration of photoacoustic imaging system. (a) The panorama of imaging system. OPO: Optical Parametric Oscillator, PD: Photodiode, BS: Beam Splitter, AH: Animal Holder, TR: Transducer, OF: Optical Fiber. RC: resin chamber (b) Close-up view of the resin chamber. Line-shaped light was delivered to the imaging object at horizontal plane, and the focused transducer was set exactly at the same horizontal plane to obtain the highest sensitivity of ultrasound detection.
A custom built line focused ultrasound transducer (5M center frequency, 80% relative bandwidth, and 40mm focus length; Doppler, Guangzhou, China) was positioned along the wall of an 80mm diameter cylindrically-shaped resin chamber, which was mounted on a rotator stage (RAP125, Zolix, China). An animal holder was set coaxially with the chamber. During an experiment, the animal was held upright while its head was attached to the head mount (AH2). The animal was held in a standing position and the resin chamber was rotated with the rotator. The line shaped light beams were delivered to the imaging object at horizontal plane, and the focused transducer was set exactly at the same plane to obtain the highest sensitivity for ultrasound detection (Fig. 2(b)). The transducer was rotated at a step size of 2°, allowing signal collection at 180 locations for one imaging experiment. Five times of signal averaging was applied and the whole scanning time was about 45s. To achieve multi-slice imaging, the head mount (AH2) was mounted on a Z-axis step motor with a scanning interval of 500 μm. The signal was amplified by a custom-made pre-amplifier with 60dB gain coupled with a band filter of 500kHz-10MHz. NI 5122 (NI USA) digital card was used for signal acquisition with a sampling frequency of 100MHz. A LabVIEW program controlled the PAT system.
2.5 Phantom testing
Two phantom experiments were conducted to validate the performance of the system. To test the lateral resolution, two ~7 μm carbon fibers were embedded in a tissue background phantom with an optical absorption coefficient of 0.01 mm−1 and a reduced scattering coefficient of 1.0 mm−1, as shown in Fig. 3(a). To test the axial resolution of our system, two suture lines were embedded in the background phantom. Three images of suture lines were reconstructed at different positions along z-axis with 500μm interval. Figures 3(d)-3(f) show the 3 images at different z-axis positions.
Fig. 3 Phantom testing. (a) Reconstructed image of the carbon fibers. (b) Lateral profile of the selected carbon fiber. (c) The geometry of 2 suture lines. (d) Image at position 1(z = 0mm), only one suture line was observed here. (e) Image at position 2(z = −0.5mm), two suture line were observed here (f) Image at position 3(z = −1mm), one suture line was observed.
2.6 Image reconstruction method
Images were reconstructed from photoacoustic signals using a universal back-projection reconstruction algorithm [28]. Functional parameters including deoxy-hemoglobin (HbR), oxy-hemoglobin (HbO2) were obtained using the method described in previous studies [17, 18]. Briefly, for every data point in the reconstructed images, the algorithm uses least-squares to match the recorded photoacoustic spectrum to the known absorption spectra of HbR and HbO2 at two wavelengths (i.e., 760nm and 840nm in this study). This results in distribution maps of the two states of hemoglobin, which are regarded as the dominant absorbers in biological tissue. Spectral un-mixing is essentially performed on a per-pixel basis by solving the following set of linear equations:
stands for the spatial distribution of the absorbed optical energy in arbitrary units. and stand for molar extinction spectra and concentration of the corresponding tissue chromophore. represents the light fluence distribution for the wavelength range employed. For we only estimate relative variation in oxy and deoxy signal, we assume the light fluence to be constant for all wavelengths employed and the light fluence distribution was neglected in estimation process following previous study [17]. Nevertheless, it is important to notice that, tissue heterogeneity, wavelength-dependence of light attenuation may introduce errors in the estimation of oxygenation levels. Accounting for those in deep brain photoacoustic imaging is another topic of investigations.
2.7 Calculations of hematoma area and cortex thickness
Quantifications of lesion area and cortex thickness during injury process contained by csPAT and histological images were obtained using ImageJ2x (Scion Corporation, Frederick, MD, USA) combing with Matlab imaging process toolbox as routinely performed previously [35]. We manually defined the brain sections including striatum, hemisphere and cortex at all time points (day1 to day 28) in the PA images under the assistance of brain atlas [30].
2.8 Functional assessment
On days 1, 3, 7, and 28 after surgery, an investigator blinded to the group assignment evaluated the mice functional deficits by an NDS(Neurologic deficit score) system that includes six domains of body symmetry, gait, climbing, circling behavior, front limb symmetry, and compulsory circling as described in previous studies [3, 29]. Each test was graded from 0 to 4, providing a total score of 0 to 24, with higher scores indicative of greater neurologic injury [29, 31, 32].
2.9 Statistical analysis
All data are reported as mean ± SD. Data were analyzed with Student’s t test or one-way ANOVA. Linear regression was used for correlational analysis between PAT and histological estimates of haemorrhagic lesion, between PAT and NDS. P-value< 0.05 was considered statistically significant. All statistical analyses were performed with GraphPad Prism 6 (GraphPad Software, La Jolla, California).
3. Results
3.1 Phantom testing results
Theoretically, the best axial and lateral resolutions of the system are approximately 1mm and 120 μm, respectively [33]. Two phantom tests were conducted to evaluate the performance of the system. Figure 3(a) shows two carbon fibers imbedded in the background phantom for the lateral resolution evaluation where one of them is indicated by the dashed red line. To verify that the axial resolution was about 1 mm, we selected two ~200μm suture lines imbedded in the background phantom with an interval of 1mm. Figures 3(b) and 3(c) show the lateral profiles of the carbon fiber and two layers of suture line, respectively. The full widths at half maximum (FWHM) of carbon fiber was 120 μm. To clearly distinguish two layers of suture lines, we imaged the objects at three different z-axis positions and showed that to distinguish two suture lines, 1mm interval was necessary.
3.2 In vivo mouse brain imaging
To acquire mouse brain images, we take advantage of low absorption coefficient and high scattering coefficient in the NIR window. Multi-slice csPAT images given in Fig. 4 show different coronal slices of mouse brain from bregma + 0 to bregma-2 at 1mm interval. Three typical slices of the mouse brain are presented in Fig. 4(a) are shown and compared with their corresponding histological sections (Fig. 4(b)). Major brain sections at different layers including cortex, caudate-putamen (CPu), thalamus (TH), hippocampus(HP) and hypothalamus (HY) were clearly revealed in the PAT images. Clear identification of these regions allowed us to easily segment these brain areas from the surrounding tissues for statistical analysis.
Fig. 4 Three typical coronal slices of a mouse brain. (a) Three slices of PAT images. CPu: caudate-putamen, cc: corpus callosum TH: thalamus, HY: hypothalamus. HP: hippocampus (b) Histology images of the corresponding PAT slices.
3.3 Monitoring of ICH process
Bregma + 0.2 mm was chosen as the main slice to quantitatively measure the lesion area identified in the PAT images and frozen slices. Haematomal lesion area was manually delineated and measured in the PAT and histologic images. On day 1 and day 3, PAT images showed that the lesion contained a region with low signal intensities surrounded by bright rims. The signal in the high-intensity region was significantly reduced on days 7 and 28 (Fig. 5(a)). From the histologic images, blood clot was observed on days 1 and 3 and resolved over the course of 28 days (Fig. 5(b)). Both PAT and histologic images showed a progressive decrease in lesion area (percent of contralateral hemisphere; day 1: 45 ± 3.1% and 46 ± 2.1%; day3: 31 ± 4.6% and 28 ± 0.89%; day 7: 9.3 ± 4.1% and 9.4 ± 2.5%; day 28: 4.2 ± 1.2% and 5.3 ± 2.2%; Fig. 1(c), 1(d);*P<0.05 vs.day1;one-way ANOVA). Correlation of lesion area obtained from the PAT images was positive, with a Pearson correlation coefficient of 0.947 (P< 0.001; r = 0.923).
Fig. 5 Development of the haemorrhagic lesion over time assessed by PAT in comparison with the histology. (a) Representative PAT images show the haemorrhagic lesion (red outline) on 1, 3, 7, and 28 days after ICH. (b) Corresponding frozen brain slices (unstained) on 1, 3, 7, and 28 days after ICH. Different cohorts of animals were used for PAT scans and for histology. (c, d, e) Quantitative analysis shows that the time-dependent changes in ratio of total lesion area to the contralateral hemisphere area were similar when assessed by PAT and histology. (PAT: n = 3 at each time point; Histology: n = 3 at each time point. CL: contralateral hemisphere)
3.4 Development of the haemorrhagic area
Two wavelengths (760nm and 840nm) were used here to monitor the hemoglobin change in haemorrhagic area. Bregma + 0.2 mm was chosen as the main slice and a threshold value of 40% was applied here based on averaged value of pre-ICH data to enhance the abnormal region. Signal value above the threshold was defined as high value region (HVR). On day 1, a large range of oxy-HVR were observed in CPu and lateral ventricle. At the borders of the lesion, areas with high values of deoxy-HVR were observed. On day 3, oxy-HVR decreased in lesion area while deoxy-HVR became the main constituent. Oxy-HVR and deoxy-HVR decreased concurrently as the lesion area shrunk and brain structure atrophied. On day 28, oxy-HVR and deoxy-HVR in haemorrhagic area came to the normal level. These process was also observed by MRI in previous studies [3, 40]. Quantitative analysis showed time-dependent changes in ratio of oxy-HVR in main lesion area and perihemmatoma indicated by blue dashed line shown in the first image of Figs. 6(a) and 6(b) (day 1: 21 ± 1.9%; day3: 8.7 ± 1.5%; day 7: 1.7 ± 0.17%; day 28: 0.81 ± 0.21%;*P<0.05 vs.day1;one-way ANOVA) and changes in the ratio of deoxy-HVR in main lesion area to contralateral hemisphere from day 1 to day 28 (day 1: 11 ± 0.46%; day3: 13 ± 0.98%; day 7: 3.5 ± 0.60%; day 28: 0.53 ± 0.24%;*P<0.05 vs.day1;one-way ANOVA).
Fig. 6 Development of the haemorrhagic area using two wavelengths. (a) Representative oxy-signal images on day 1, 3, 7, and 28 days after ICH. (b) Representative deoxy-signal images on day 1, 3, 7, and 28 days after ICH. (c) Corresponding frozen brain slices (unstained) on 1, 3, 7, and 28 days after ICH. (d, e) Quantitative analysis shows the time-dependent changes in ratio of HVR in CPu to the contralateral hemisphere area. (CL = contralateral;*P<0.05 vs day1;one-way ANOVA).
3.5 Brain region area changes and Neurologic deficit score (NDS)
Based on PA image, we obtained the area of major structures at main slice (Bregma + 0.2mm) in the ICH brain via segmentation (Fig. 2(a), 2(b)). The relative area of the ipsilateral striatum (percent of contralateral hemisphere) increased significantly from 26.3 ± 1.7% pre-ICH to 44.6 ± 1.6% at 1 days and then 42.7 ± 0.99% and decreased to 32.7 ± 2.9% at 7 days and 25.5 ± 2.3% at 28 days(*P<0.05 vs.day1;one-way ANOVA; #P<0.05 vs.day1;one-way ANOVA). The ration of cortex thickness ration to contralateral cortex thickness decreased from 96 ± 0.67% pre-ICH to 63 ± 7.2% at 1 days and then 0.60 ± 6.9% at 3 days and 68 ± 6.9% at 7 day and 78 ± 2.1% at day 28(*P<0.05 vs.day1;one-way ANOVA). A standard NDS test was also applied here in our study. The NDS decreased from day 1 (10.2 ± 3.3) to day 3 (10.2 ± 3.3), day 7 (7.6 ± 2.3) and day 28 (3.5 ± 0.7; all P< 0.001; Fig. 7(d)). Correlation between NDS and the PAT-identified lesion area was positive, with a Pearson correlation coefficient of 0.9367 (P< 0.001; Fig. 7(e)).
Fig. 7 Neurologic deficit score (NDS) correlates with brain region area changes assessed by PAT. (a) Major structures in the ICH brain in PA image (white line indicated the cortex thickness). (b) Quantitative analysis of PA image showed that the ratio of ipsilateral striatum to contralateral hemisphere area increased significantly at day 1 and day 3, then progressively decreased at days 7 and 28 (CL: contralateral hemisphere). (c)The cortex thickness also decreased significantly after ICH then increased at day 28 (d) Neurologic deficit progressively decreased from day 1 to day 28 after ICH (n = 6 mice). (e) Scatter plot shows positive correlation between lesion area on PAT and NDS (n = 3 mice).
4. Discussion and conclusions
In this study, for the first time, we applied cross-sectional PAT to investigate injury progression after intracerebral haemorrhage. By utilizing the NIR window we achieved deep brain imaging of small animals in high resolution. In the results presented here, a series of pathological alterations such as hematoma, mid-shift (superior sagittal sinus (SSS) moved from the middle to one hemisphere [34]), and HbR and HbO2 changes in the surrounding hemorrhagic area induced by hematoma compression after injection [34, 35] were clearly observed from our PA images.
It was noted that the lesion area measured by csPAT was similar to those reported in previous studies [34]. PAT and frozen slices both showed that the lesion area increased from day 1 to day 3 and decreased from day 3 to day 28. Mid-shift was shown during the whole progression. Correlation between lesion areas as determined by PAT versus histological examination of cryo-slices was shown here. These results suggest that csPAT can be used to provide quantitative estimates of haemorrhagic lesion like MRI. Also, correlation between PAT-determined lesion area and NDS were fund in our research, indicating that PAT might be useful for evaluating the outcome of ICH and efficacy of treatments.
Hemoglobin change in brain hematoma is another important development process during ICH process, supported by previous fMRI studies [29]. Clinically, ICH is caused by rupture of a tiny artery, which leads to formation of a haematoma or blood clot in the lesion core, with tissues pressed and distorted in the periphery [3, 36]. At the beginning, with the rupture of a tiny artery, the main hemoglobin formation was HbO2 in the lesion core [35]. With the process of haematoma, hemoglobin protein concentration increased, and the main form of hemoglobin changed from HbO2 to HbR [35], as shown in PAT images. As the injury progressed, the HbR turned into MbR (Methemoglobin) from the outer layer of hematoma and then gradually to the core area [35], so the deoxy-signal decreased from day 3 to day 28.
Brain tissue loss is common in ICH models [29, 32]. Histology results in previous studies showed that tissue loss and brain atrophy occured in several brain regions, including cortex [29] and striatum [37]. On csPAT images, we measured the area of striatum and thickness of cortex to determine which brain regions changed because of long-term tissue loss after ICH. Our findings were similar to those of a previous MRI study in rats and mice [3, 29].
In summary, our current study shows that cross-sectional PAT is capable of noninvasively monitoring the injury process of collagenase induced ICH in vivo. Traditional PAT cannot afford deep brain imaging which is more important in ICH for detecting hematoma area and computing tissue loss. Cornal plane imaging demonstrated hematoma that were more intuitive compared with the histologic image. Besides, a standard functional assessment applied here according to NDS made our work reliable and rigid. These results were not shown in our previous work [17]. There are several limitations in our work indeed. Data acquisition speed of this system is relative low, which can be improved by using a transducer array [16]. The use of a faster laser can also improve the temporal resolution of the system. Nonetheless, this study suggests that csPAT offers a powerful tool to evaluate therapeutic interventions in preclinical ICH models.
Disclosures
The authors declare that there are no conflicts of interest related to this article.
References and links
1. T. Steiner, R. Al-Shahi Salman, R. Beer, H. Christensen, C. Cordonnier, L. Csiba, M. Forsting, S. Harnof, C. J. Klijn, D. Krieger, A. D. Mendelow, C. Molina, J. Montaner, K. Overgaard, J. Petersson, R. O. Roine, E. Schmutzhard, K. Schwerdtfeger, C. Stapf, T. Tatlisumak, B. M. Thomas, D. Toni, A. Unterberg, and M. Wagner, “European Stroke Organisation (ESO) guidelines for the management of spontaneous intracerebral hemorrhage,” Int. J. Stroke 9(7), 840–855 (2014). [PubMed]
2. C. J. van Asch, M. J. Luitse, G. J. Rinkel, I. van der Tweel, A. Algra, and C. J. Klijn, “Incidence, case fatality, and functional outcome of intracerebral haemorrhage over time, according to age, sex, and ethnic origin: a systematic review and meta-analysis,” Lancet Neurol. 9(2), 167–176 (2010). [PubMed]
3. J. Yang, Q. Li, Z. Wang, C. Qi, X. Han, X. Lan, J. Wan, W. Wang, X. Zhao, Z. Hou, C. Gao, J. R. Carhuapoma, S. Mori, J. Zhang, and J. Wang, “Multimodality MRI assessment of grey and white matter injury and blood-brain barrier disruption after intracerebral haemorrhage in mice,” Sci. Rep. 7, 40358 (2017). [PubMed]
4. M. M. McDowell, C. P. Kellner, S. M. Barton, C. B. Mikell, E. S. Sussman, S. G. Heuts, and E. S. Connolly, “The role of advanced neuroimaging in intracerebral hemorrhage,” Neurosurg. Focus 34(4), E2 (2013). [PubMed]
5. J. Wang, “Preclinical and clinical research on inflammation after intracerebral hemorrhage,” Prog. Neurobiol. 92(4), 463–477 (2010). [PubMed]
6. 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). [PubMed]
7. O. I. Alomair, I. M. Brereton, M. T. Smith, G. J. Galloway, and N. D. Kurniawan, “In vivo high angular resolution diffusion-weighted imaging of mouse brain at 16.4 Tesla,” PLoS One 10(6), e0130133 (2015). [PubMed]
8. T. Kim, R. Zhou, M. Mir, S. D. Babacan, P. S. Carney, L. L. Goddard, and G. Popescu, “White-light diffraction tomography of unlabelled live cells,” Nat. Photonics 8, 256–263 (2014).
9. N. G. Horton, K. Wang, D. Kobat, C. G. Clark, F. W. Wise, C. B. Schaffer, and C. Xu, “In vivo three-photon microscopy of subcortical structures within an intact mouse brain,” Nat. Photonics 7(3), 205–209 (2013). [PubMed]
10. F. Zhang, L. P. Wang, M. Brauner, J. F. Liewald, K. Kay, N. Watzke, P. G. Wood, E. Bamberg, G. Nagel, A. Gottschalk, and K. Deisseroth, “Multimodal fast optical interrogation of neural circuitry,” Nature 446(7136), 633–639 (2007). [PubMed]
11. V. Ntziachristos, “Going deeper than microscopy: the optical imaging frontier in biology,” Nat. Methods 7(8), 603–614 (2010). [PubMed]
12. C. Darne, Y. Lu, and E. M. Sevick-Muraca, “Small animal fluorescence and bioluminescence tomography: a review of approaches, algorithms and technology update,” Phys. Med. Biol. 59(1), R1–R64 (2014). [PubMed]
13. P. Beard, “Biomedical photoacoustic imaging,” Interface Focus 1(4), 602–631 (2011). [PubMed]
14. J. Tang, L. Xi, J. Zhou, H. Huang, T. Zhang, P. R. Carney, and H. Jiang, “Noninvasive high-speed photoacoustic tomography of cerebral hemodynamics in awake-moving rats,” J. Cereb. Blood Flow Metab. 35(8), 1224–1232 (2015). [PubMed]
15. J. Yang, D. Wu, Y. Tang, and H. Jiang, “Photoacoustic microscopy of electronic acupuncture (EA) effect in small animals,” J. Biophotonics 10(2), 217–223 (2017). [PubMed]
16. L. Li, L. Zhu, C. Ma, L. Lin, J. Yao, L. Wang, K. Maslov, R. Zhang, W. Chen, and J. Shi, “Single-impulse panoramic photoacoustic computed tomography of small-animal whole-body dynamics at high spatiotemporal resolution,” Nature Biomedical Engineering 1, 0071 (2017).
17. X. L. Deán-Ben and D. Razansky, “Functional optoacoustic human angiography with handheld video rate three dimensional scanner,” Photoacoustics 1(3-4), 68–73 (2013). [PubMed]
18. D. Wu, J. Yang, G. Zhang, and H. Jiang, “Noninvasive in vivo monitoring of collagenase induced intracerebral hemorrhage by photoacoustic tomography,” Biomed. Opt. Express 8(4), 2276–2286 (2017). [PubMed]
19. I. Olefir, E. Mercep, N. C. Burton, S. V. Ovsepian, and V. Ntziachristos, “Hybrid multispectral optoacoustic and ultrasound tomography for morphological and physiological brain imaging,” J. Biomed. Opt. 21(8), 86005 (2016). [PubMed]
20. L. Xi and H. Jiang, “High resolution three-dimensional photoacoustic imaging of human finger joints in vivo,” Appl. Phys. Lett. 107, 063701 (2015).
21. L. Xi and H. Jiang, “Integrated photoacoustic and diffuse optical tomography system for imaging of human finger joints in vivo,” J. Biophotonics 9(3), 213–217 (2016). [PubMed]
22. N. C. Burton, M. Patel, S. Morscher, W. H. Driessen, J. Claussen, N. Beziere, T. Jetzfellner, A. Taruttis, D. Razansky, B. Bednar, and V. Ntziachristos, “Multispectral opto-acoustic tomography (MSOT) of the brain and glioblastoma characterization,” Neuroimage 65, 522–528 (2013). [PubMed]
23. X. L. Deán-Ben and D. Razansky, “Adding fifth dimension to optoacoustic imaging: volumetric time-resolved spectrally enriched tomography,” Light Sci. Appl. 3, e137 (2014).
24. S. Gottschalk, T. F. Fehm, X. L. Deán-Ben, V. Tsytsarev, and D. Razansky, “Correlation between volumetric oxygenation responses and electrophysiology identifies deep thalamocortical activity during epileptic seizures,” Neurophotonics 4(1), 011007 (2017). [PubMed]
25. N. C. Burton, M. Patel, S. Morscher, W. H. Driessen, J. Claussen, N. Beziere, T. Jetzfellner, A. Taruttis, D. Razansky, B. Bednar, and V. Ntziachristos, “Multispectral opto-acoustic tomography (MSOT) of the brain and glioblastoma characterization,” Neuroimage 65, 522–528 (2013). [PubMed]
26. P. R. Krafft, W. B. Rolland, K. Duris, T. Lekic, A. Campbell, J. Tang, and J. H. Zhang, “Modeling Intracerebral Hemorrhage in Mice: Injection of Autologous Blood or Bacterial Collagenase,” J. Vis. Exp. 67(67), e4289 (2012). [PubMed]
27. B. R. Masters and L. V. Wang, Photoacoustic Imaging and Spectroscopy (CRC Press, 2009).
28. M. Xu and L. V. Wang, “Universal back-projection algorithm for photoacoustic computed tomography,” Phys. Rev. E Stat. Nonlin. Soft Matter Phys. 71(1 Pt 2), 016706 (2005). [PubMed]
29. C. L. MacLellan, G. Silasi, C. C. Poon, C. L. Edmundson, R. Buist, J. Peeling, and F. Colbourne, “Intracerebral hemorrhage models in rat: comparing collagenase to blood infusion,” J. Cereb. Blood Flow Metab. 28(3), 516–525 (2008). [PubMed]
30. G. Paxinos and C. Watson, The rat brain in stereotaxic coordinates, (Academic Press, 1998).
31. J. Wang and S. Doré, “Heme oxygenase-1 exacerbates early brain injury after intracerebral haemorrhage,” Brain 130(Pt 6), 1643–1652 (2007). [PubMed]
32. W. Zhu, Y. Gao, C. F. Chang, J. R. Wan, S. S. Zhu, and J. Wang, “Mouse models of intracerebral hemorrhage in ventricle, cortex, and hippocampus by injections of autologous blood or collagenase,” PLoS One 9(5), e97423 (2014). [PubMed]
33. A. Ng and J. Swanevelder, “Resolution in ultrasound imaging,” Continuing Education in Anaesthesia, Critical Care Pain 11, 186–192 (2011).
34. G. Xi, R. F. Keep, and J. T. Hoff, “Mechanisms of brain injury after intracerebral haemorrhage,” Lancet Neurol. 5(1), 53–63 (2006). [PubMed]
35. L. A. Hayman, K. H. Taber, J. J. Ford, and R. N. Bryan, “Mechanisms of MR signal alteration by acute intracerebral blood: old concepts and new theories,” AJNR Am. J. Neuroradiol. 12(5), 899–907 (1991). [PubMed]
36. M. Li, R. M. Akhavan-Sharif, R. M. Friedlander, R. Du, and R. Thiex, “What sequences on high-field MR best depict temporal resolution of experimental ICH and edema formation in mice?” J. Biomed. Biotechnol. 2012, 961461 (2012). [PubMed]
37. K. Karki, R. A. Knight, Y. Han, D. Yang, J. Zhang, K. A. Ledbetter, M. Chopp, and D. M. Seyfried, “Simvastatin and atorvastatin improve neurological outcome after experimental intracerebral hemorrhage,” Stroke 40(10), 3384–3389 (2009). [PubMed]
