Open Access
Add to BibSonomy
Abstract
In our previous studies, we have demonstrated the feasibility of characterizing intestinal inflammation and fibrosis using endoscopic photoacoustic imaging. Purposed at te clinical translation of the imaging technology, we developed a photoacoustic/ultrasound imaging probe by integrating a miniaturized ultrasound array and an angle-tipped optical fiber in a hydrostatic balloon catheter. When collapsed, the catheter probe may potentially be compatible with a clinical ileo-colonoscope. In addition, the flexible surface of the hydrostatic balloon allows for acoustic coupling at the uneven surfaces of the gas-filled intestine. Tissue phantom studies show that the catheter probe possesses an imaging penetration of at least 12 mm. Experiments with a rabbit model in vivo validated the probe in differentiating normal, acute and chronic conditions in intestinal obstruction.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Crohn’s Disease is a chronic gastrointestinal disease that commonly causes obstructing intestinal strictures due to a mixture of inflammation, fibrosis, and muscular hypertrophy [1,2]. The standard diagnostic procedure, ileocolonoscopic biopsy, removes small pieces of mucosa using biopsy forceps through the instrument channel of a colonoscope. Histopathology of the biopsy samples reveals inflammatory cells and distorting structural damage to the epithelium as markers of acute and chronic inflammation. However, mucosal biopsies only sample the superficial epithelial layer of the intestine and cannot assess the biology of fibrostenotic disease, which occurs in the deeper layers of the bowel wall [3].
In our previous studies, we have demonstrated the feasibility of transabdominal and endoscopic photoacoustic (PA) imaging through the full thickness of intestinal strictures in human subjects and animals [4]. Compared to established imaging modalities such as magnetic resonance imaging, X-ray computed tomography and ultrasound, PA imaging possesses the unique advantage of quantifying hemoglobin and collagen content [5–8] as measurements of inflammation and fibrosis, respectively. Although transabdominal PA imaging coupled with US is attractive for quick clinical translation, the approach produces images at limited resolution and is challenged by optical attenuation through the abdominal wall, especially in obese patients who have often been treated with corticosteroids. Approaching this problem via endoscopic imaging, on the other hand, brings the imaging elements close to the region of interest, and, therefore, renders images with higher resolution and minimizes the quantification uncertainty caused by wavelength-dependent optical attenuation [5].
Delivering a miniaturized imaging device to the intestinal stricture through the instrument channel of a colonoscope poses minimal interruption to the established ileocolonoscopy procedure and may substantially improve the diagnostic accuracy of this procedure. Miniaturized endoscopic PA-US imaging has been investigated by a few groups [9–11], including us [5]. These studies relied on the rotational or linear scanning of a single focused US transducer. Although focused US transducers provide high resolution within the focal zone, their limited depth of field undermines their ability to image the muscle layer in the bowel, which can extend up to 12 mm from the epithelial surface in stenotic bowel wall [12,13]. Furthermore, most previous endoscopic PA-US imaging probes employed a rigid surface due to the complicated internal scanning mechanisms [9–11]. This rigid design works well in an intravascular imaging scenario where surrounding liquid blood facilitates acoustic coupling [10]. However, for acoustic coupling with the uneven surfaces of the (often) gas-filled intestine, a flexible and adaptive probe surface is necessary. Hydrostatic balloon catheters are frequently used during ileocolonoscopy procedures for dilating the lumen of short (<5 cm) obstructing strictures [14,15]. When inflated with water, the ballon surface makes full contact with the inner lumen of an intestinal stricture and forms a perfect acoustic coupling medium for ultrasound.
In this study, we demonstrate the utility of a prototype endoscopic PA tomography and US imaging balloon catheter that integrates a side firing optical fiber and a commercial miniaturized US transducer array. Such combination, with a limited sacrifice of imaging resolution, allows for the desirable penetration in the intestinal imaging scenario. The performance of the catheter was examined in phantoms and in an animal model of intestinal inflammation and fibrosis.
2. Methods and materials
2.1 Balloon catheter
As shown in Fig. 1, the probe integrates a commercial US array (Acunav 8F, Siemens, Seattle, WA) and an angled-tip optical fiber inside a medical balloon catheter. The Acunav is a 64-element miniaturized linear US array with a pitch of 110 µm and a bandwidth of 5-10 MHz. The emission tip of the optical fiber (0.39 NA, 800 µm core, Thorlabs, Newton, NJ, USA) was polished at a 23° angle. A 3-cm-long, semi-rigid tube (144-0030, Nordson Medical, Ohio, USA) maintains the relative positions of the US array and the emission end of the side-firing fiber. The functional elements are exposed through a 1-cm-long window along the tube. The whole structure is encapsulated by a balloon catheter consisting of a hydrostatic balloon (10 mm dimeter, 10002000BB, Nordson Medical, Ohio, USA) and a heat-shrunk catheter body (103-0663, Nordson Medical, Ohio, USA). The gap between the catheter body and the inner elements serves as the water conduit for balloon inflation and deflation. The coupling end of the catheter body connects to a T adapter for water injection. A 3 way stopcock (Masterflex Fitting, Cole-Palmer, Vernon Hills, IL) allows for the injection of water from a syringe or a hydrostatic pump and for the withdrawal of air or water from the balloon catheter. Figure 2 shows the photos of the probe. The probe has a diameter of approximately 3.7 mm with the hydrostatic balloon collapsed.
Fig. 1. The Assembly of the PA-US balloon catheter. The structures marked with green boxes on top are magnified in the bottom. 1) Side-firing fiber optics with 800 µm core. 2) Illumination delivered by the side-firing fiber optic. 3) US array. 4) Semi-rigid tube to hold the relative positions of the side-firing fiber and the US transducer. 5) Hydrostatic balloon. 6) Three-way stopcock. 7) T adapter. 8) Hydrostatic pump.
Fig. 2. Photos of the distal end of catheter probe. (a) and (b) are deflated and inflated catheter probe, respectively. (c) The catheter probe with illumination emission. Scalebars are 10 mm.
At PA imaging mode, the angled-tip optical fiber delivers the output of an optical parametric oscillator (OPO) pumped by the second harmonic of an Nd:YAG laser (Phocus Mobile, OPOTEK, Santa Clara, CA) with pulse width of 8 ns and repetition rate of 10 Hz. The OPO has emission range of 690-950 nm and 1200-2400 nm, with a peak output energy of 150 mJ at approximately 720 nm. Figure 3(a-b) illustrates the emission pattern of the optical fiber. As reported in a previous study [16], optical energy is split into refraction and reflection beams at the angled emission surface. With the purpose of maximizing the reflected energy, and considering the fiber core, cladding and water reflection indices of 1.45, 1.40 and 1.33, the fiber tip was polished at 23° (=90°-arcsin(nwater/ncore)), which is the total internal reflection angle of optical energy at the fiber core/water interface. Figure 3(a-b) shows that the majority of the optical energy is projected into the reflection direction, although a small amount of optical energy leaks into the diffraction direction. As shown in Fig. 2(c) and Fig. 3(a), the forward-going optical energy does not illuminate the tissue volume covered by the US array. The reflected beam, exiting the fiber through the fiber core and cladding, possesses a skewed gaussian energy distribution in the longitudinal direction, and expands in the azimuthal direction due to the cylindrical interface between the cladding and water [16]. The projection angles at the interfaces and the beam profiles at the balloon surface (4 mm from the fiber) were calculated, as shown in Fig. 3(a) and (b), respectively. The side-firing beam delivers an average optical density of 4 mJ/cm2 to an area of approximately 1.5 cm2 at the balloon surface. Considering the extensive expansion of the beam profile in the azimuthal direction, we assume that the optical energy distribution is uniform within the narrow elevational aperture of the US and array. Therefore, the optical energy distribution within the field-of-view is simulated in 2D using a MATLAB toolbox, NIRFAST [17], for compensating optical attenuations in all cases in this study. The optical energy distribution in Fig. 3 (d) was calculated using optical properties of 1% intralipid water solution at 720 nm, i.e. µa= 0.0027 mm-1 and µs’= 1 mm-1, where µa and µs’ are the absorption and the reduced scattering coefficients, respectively.
Fig. 3. Optical illumination for PA imaging and compensation to the nonuniform optical energy distribution within the imaging plane. (a) Geometric derivation of the emission orientation of the optical energy. We used the total internal reflection angle at the fiber core/cladding interface (15.4o = arcsin(ncore/ncladding)) to find the emission range. (c) Optical energy distribution within the imaging plane at 720 nm calculated based on the skewed gaussian emission profile derived in (a). (d) Top view of the imaging phantom made of polyethylene fibers. Panels (e) and (f) are the PA images of the phantom without and with compensation for the optical energy distribution, respectively. (g) Cross-sectional plots in (e-f). (h) Estimated point-spread-function. (i) The axial and lateral point-spread-functions along the red and blue lines in (h), respectively. Full-width-half-maxima of the point-spread-functions are 112 µm and 181 µm, respectively.
The generated PA signals, as well as pulse-echo US signals, were acquired by an US platform (Vantage 256, Verasonics, Kirkland, WA). Images of both modalities were reconstructed with the delay-and-sum method. Figure 3 (e-f) shows the reconstructed PA images before and after compensated by optical energy distribution. The cross-sectional plot in Fig. 1(g) shows that the variation of the reconstruction fibers in the compensated image (standard deviation of peak magnitudes = 14%) is less than that in the uncompensated image (standard deviation of peak magnitudes = 21%).
2.2 Examining the probe performance with imaging phantoms and ex vivo tissue
The sensitivity distribution of the imaging probe was initially examined using an imaging phantom. Nine 200 µm polyethylene fibers were evenly distributed perpendicular to the imaging plane with separations of approximately 2 mm within the imaging plane, as shown in Fig. 3(c). As our goal was to mimic the diffusive optical properties of biological tissue, the fibers were submerged in a background material of 1% intralipid (Baxter, Deerfield, IL) in 10% porcine gelatin (Sigma-Aldrich, Cleveland, OH) by weight. Figure 3(d) is the calculated optical energy distribution. The contrast to noise ratio, calculated as the ratio of the averaged peak intensity of the reconstructed polyethylene fibers over the averaged background pixel value, is 8.0. After performing compensation adjustments on the PA image in Fig. 3(e) using Fig. 3(d), Fig. 3(f) show more consistent contrasts at all targets. Considering the small field of view and relative uniform reconstruction sizes of the polyethylene fibers, we used a MATLAB built-in function, deconvblind to estimate the point spread function of the PA imaging modality. Figure 3(h) shows that the estimated axial and lateral full-width-half-maxima of the estimated point spread function are 112 µm and 181 µm, respectively.
The catheter probe was also used to examine a tissue phantom to measure imaging penetration. As shown in Fig. 4, the tissue phantom consists of 2 pieces of porcine intestines rolled into a tube shape and fixed in 10% porcine gel. The thickness of the tissues was approximately 12 mm, which is comparable to that of high-grade stenotic bowel wall thickness in humans [12,13]. The balloon catheter was inserted into the 12-mm-diamter opening of the phantom and rotated by a step motor for 360 degrees with a step size of 0.48 degree. With the purpose of qualitatively examining the capability of the imaging method in differentiating the tissue components, 800 and 1310 nm wavelengths were selected for illumination, targeting the strong optical absorption of hemoglobin and collagen content in the tissue samples. We did not select 720 nm as we did in our previous study [5] because oxygenated and deoxygenated hemoglobin possess similar optical absorption coefficients at 800 nm and we only wish to measure the total hemoglobin content in this study. At each step, photoacoustic signals were acquired with two consecutive illumination pulses at the two wavelengths with 0.1 second separation. The light distribution at 800 and 1310 nm were calculated using the approximated optical properties of intestinal tissue in our previous studies [5], i.e. µa= 0.42 mm-1 and µs’= 1.18 mm-1 at 800 nm and µa= 1.31 mm-1 and µs’= 0.5 mm-1 at 1310 nm. The relative concentrations of hemoglobin and collagen content are resolved using the spectroscopic decoupling method described in our previous study [5]. A three-dimensional image was rendered by combining the 2D images at all steps.
Fig. 4. Tissue phantom study. (a) Optical absorption profiles of hemoglobin (both oxy- and de-oxy hemoglobin) and collagen [6]. (b) 2 pieces of porcine intestinal tissue. (c) The 2 pieces of tissue are stacked to form a thickness of 12 mm. The samples were rolled and sealed in porcine gelatin to mimic the geometry of endoscopic imaging.
2.3 Examining the performance of the balloon catheter probe in animal models in vivo
The laboratory animal protocol and ethics for this study was approved by the Institutional Animal Care and Use Committee at the University of Michigan, Ann Arbor. New Zealand white rabbits of both sexes (Covance, Kalamazoo, MI, USA or University of Michigan, ULAM Breeding Colony, Ann Arbor, MI, USA) were used in this study. During all procedures, the animals were maintained under general anesthesia during PA-US imaging using 4-5% inhaled isoflurane. Vital signs and anesthesia depth were continuously monitored during anesthesia. To allow histologic assessment of the imaged bowel wall, animals were euthanized by barbiturate overdose in compliance with our institutional UCUCA policy and the AVMA Guidelines on Euthanasia.
2.3.1 Rabbit model
The rabbit model [18] used in our previous study [6] was employed to examine the performance of the catheter probe in vivo. In brief, to induce acute inflammation, a single intra-rectal dose of 40 mg TNBS in 25% to 30% ethanol was delivered 7 cm into the rectum/distal colon using a 10 French size catheter. To induce intestinal fibrosis, multiple escalating intra-rectal doses of TNBS (from 40 to 80 mg) with 2 to 4 week healing intervals were used to replicate the cycles of inflammation and tissue repair that trigger chronic intestinal fibrosis as we and others have observed in the rat TNBS model [19–21]. 9 normal rabbits, 8 acute rabbits and 8 chronic rabbits were used.
2.3.2 Imaging procedures
Before the PA scanning, the colon was first visually examined from the rectum to ∼12 cm proximal using a bronchoscope (CF-HQ190L, Olympus, Center Valley, PA) which accommodates the narrow lumen of the rabbit intestine, to determine the extent of disease. Video recordings of the bowel and representative still images were collected. After the bronchoscope was removed, the balloon catheter was inserted into the colon and positioned at regions corresponding to the visually observed damage, then fully dilated to achieve acoustic coupling with the intestinal wall. Lubricant (Surgilube, HR Pharmaceuticals Inc., York, PA) was used for insertion of both the bronchoscope and the balloon catheter to avoid damaging the fragile colon surface. Similar to the tissue sample experiments, laser illumination at 800 and 1310 nm was used. The laser was then switched to a wavelength-swapping mode where photoacoustic images at both wavelengths, and an US image, was taken. For each position, we acquired 100 frames in 10 seconds.
2.4 Data quantification and statistics
We used the same method in the phantom study in 2.2 to resolve the contribution of blood and collagen content at each pixel [5,22]. The bowel wall region identified in US images was selected as the region of interest (ROI), and the average pixel intensity in the ROI was used as the quantification result. The statistical analysis for differentiating the normal, acute, and chronic conditions were performed using the two-tailed t-test function in MATLAB (2017, Mathworks, Natick, MA, USA). The null hypothesis is that the three conditions (healthy, acutely inflamed, chronically fibrotic) cannot be differentiated using PA measurements.
2.5 Histology
Proximal (untreated, i.e. normal) and distal (treated) colons of the euthanized rabbits were harvested and processed for histology. Hematoxylin and eosin (H&E) staining and Masson’s trichrome staining were performed by the University of Michigan Cancer Center Histology and Immunoperoxidase Lab (Ann Arbor, MI, USA) and McClinchey Histology Lab (Stockbridge, MI, USA), respectively. Digital photomicrographs of proximal and distal colon sections were captured using an Olympus BX51 microscope at the University of Michigan Microscopy and Image Analysis Laboratory (Ann Arbor, MI, USA).
3. Results
3.1 Tissue phantom study
Figure 5 shows the US and PA images acquired at a rotation step in Fig. 5(a-c) and the 3D image of the tissue phantom (Fig. 5(c)). The blood and collagen content are mainly shown in the PA images acquired at 800 and 1310 nm, respectively. After the spectroscopic decoupling, the two components are encoded in pseudocolors in the 3D image. PA imaging has successfully penetrated through the tissue phantom with a thickness of 12 mm. The collagen/lipid rich layers and the blood spots were successfully differentiated. The blood clot with a large extension in Fig. 4(d) has also been identified.
Fig. 5. Images of the tissue phantom. (a) Ultrasound image. (b) and (c) are PA images at one rotational step at 800 and 1310 nm, respectively. (d) 3D image of the tissue phantom combining spectrally unmixed PA images and all rotational steps.
3.2 Rabbit imaging in vivo
Figure 6 shows representative endoscopic images of rabbit colons from normal (untreated), acute (single TNBS treatment), and chronic (multiple cycles of TNBS treatment and healing). In Fig. 6(a), the normal colon shows clear vasculature at the inner surface of the colon. Figure 6(b) acquired from acutely inflamed rabbit colon shows observable ulceration and the loss of superficial vasculature. Figure 6(c) shows a typical colon in chronically fibrotic rabbit colon, which also has the loss of vasculature and pale color seen in the acutely inflamed colon, but no ulcers, demonstrating mucosal healing after several rounds of inflammation. In the colonoscopy images, only the superficial mucosal surface of the colon can be observed.
Fig. 6. Endoscopic images of the rabbit model. (a) Normal control. (b) Animal with acute inflammation and ulceration. (c) Chronic animal after multiple TNBS treatments and cycles of inflammation and mucosal healing.
Figure 7 shows the representative US and PA images of the rabbit colon. The gray scale US images of all three conditions span the entire depth of the colon from the mucosal lumen to the outer serosal surface. Colonic wall thickening occurred in both acutely and chronically inflamed rabbit colons. In addition, the chronic rabbit shows slightly more uniform texture in the US images, probably due to mucosal healing and scarring. The hemoglobin image acquired form the normal animal shows observable signals at the surface, which is related to the vascularity observed in the endoscopic images. The bright spot on the surface may be the cross-section of a large vessel. The strong hemoglobin signals at the inner surface of the acute animal agree with the fact that ulceration involves active inflammation and bleeding. The hemoglobin signals in the deep layers of the colon in chronically treated rabbits have not been observed in the endoscopic images. The collagen images only show strong signals in the chronic animals, which agrees with the fact that multiple treatments lead to the development of chronic fibrosis in the colon. The overlaid hemoglobin and collagen images show the relative intensity and depth of the observed signals. Similar to our previous in vivo study [5], the findings were confirmed by histopathology in Fig. 7.
Fig. 7. Representative images. The images were normalized by the highest pixel values seen in each of the modalities. The blue and green arrows in the US images mark the mucosal and serosal surfaces of the colons, respectively. In the histology image of the chronic animal, black arrows denote areas of pathological collagen scarring in the mucosa. Scale bar = 200 µm.
3.3 Statistical results
Figure 8 shows the statistics of the quantitative measurements from all animals. An approximately 10-fold increase in hemoglobin signal intensity was observed in the acute animals compared to that in the normal animals [Fig. 8 (a) p < 0.05]. The chronic animals demonstrate approximately 2-fold hemoglobin signal increase compared to the normal ones [Fig. 8 (a) p < 0.05]. The chronic animals have an approximately 2-fold collagen signal intensity compared to the normal and acute ones [Fig. 8 (b), p < 0.05]. These results agree with the conclusions of our previous studies [4,5], i.e. the normal and acute conditions can be differentiated with hemoglobin signal changes, and the chronically fibrotic condition can be differentiated from normal and acute ones with collagen signal changes.
Fig. 8. Statistics of the relative content of hemoglobin and collagen resolved from the PA measurements. Blue crosses are the relative contents. The red circle represents the mean values of the corresponding groups. Blue error bars represent the standard deviations. The p-values were derived from the t-tests between the groups marked by the brackets.
4. Discussion
These experimental results validated the feasibility of implementing a balloon catheter for endoscopic PA and US imaging. The flexible hydrodynamic balloon filled with water served effectively as an adaptive coupling medium. The side firing optical fiber, in combination with the miniaturized US array, showed sufficient penetration in imaging phantoms and desirable sensitivity to the tissue components of interest in rabbit colons. Instead of mechanical scanning with a single element transducer, the catheter probe forms images by beamforming with multiple transducers. This design substantially improves the maximum frame rate of the imaging system. Targeted at simplicity and reliability for clinical translation, the probe does not have rotation mechanism for 3D scanning. We will only take 2D images by rotating the imaging plane to the region-of-interest when the catheter probe is integrated into colonoscopes. The balloon catheters are assembled under normal atmosphere. Therefore, small air bubbles are unavoidable, which may interfere with the acoustic energy propagation. Currently we take advantage of gravity to avoid these bubbles, which rise upward in the balloon, and only look downward in the imaging procedures and reposition the animals if necessary. Applying vacuum within the catheter during assembly may resolve this minor issue in the future.
The bronchoscope was used to observe three disease conditions at the epithelial surface of the animal colons. However, due to its small dimensions, the instrument channel of the bronchoscope does not allow the passage of the imaging catheter. The PA images in Fig. 7 were taken after the bronchoscope was removed. It is ideal that the catheter probe can be positioned with the guidance of the endoscope, particularly in the human terminal ileum in Crohn's disease. The diameter of the prototype catheter probe in this study is slightly larger than the instrument channel of an adult colonoscope (CF-HQ190L, Olympus), where the instrument channel has a diameter of 3.7 mm. In addition, the length of the miniature US array (1.2 m) is too short to run through the instrument channel (1.68 m). We are working on a customized US array with a longer cable and without the outer catheter body [the blue protective layer of the US array shown in Fig. 2(a)]. Such an US array will allow for a slimmer design by integrating the US array and a bare optical fiber into a balloon catheter. The small diameter of this catheter probe will also allow rotation of the catheter for imaging the full circumference of the intestinal obstruction.
Since the diameter of the rabbit bowel showed limited variations and the low durometer material allowed for minor adjustments of the balloon diameter by manipulating the internal hydrostatic pressure, only one balloon size was used in this study. However, when imaging intestinal obstructions in human subjects, catheter probes with a wide range of balloon sizes may be prepared. An appropriate size will be selected based on direct observation through the colonoscope or pre-procedure computed tomography. The diameter of the fully deployed balloon will be measured in the US images. The illumination distribution at the balloon surface will be calculated for optical energy compensation in PA images.
The contacting pressure exerted by the hydrostatic balloon may partly collapse the vasculature at the intestine wall and introduce uncertainty to the quantitative measurements of hemoglobin. The compliance of the intestine to hydrostatic balloon pressure can be measured using a pressure gauge. The pressure gauge with the hydrostatic pump in the current system has limited sensitivity. A more sensitive pressure gauge will be added to the water injection route for improved measurement accuracy.
5. Conclusion
We developed a prototype PA-US dual imaging balloon catheter that can potentially be compatible with commercially available endoscopes and endoscopic procedures. The probe has demonstrated the ability to differentiate the molecular components in a tissue phantom with thickness comparable to human intestinal strictures and characterize the molecular content of inflammatory and fibrotic lesions in rabbit colons in vivo. The unique flexible design and small dimension of the probe allow for quick clinical translation and more accurate diagnosis.
Funding
National Institute of Diabetes and Digestive and Kidney Diseases (R01DK125687); Crohn's and Colitis Foundation (581584); National Cancer Institute (R37CA222829).
Acknowledgments
This research was supported in part by NIH R37CA222829, NIH R01DK125687 and Crohn’s and Colitis Foundation, senior research award 581584.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. G. R. Lichtenstein, A. Olson, S. Travers, R. H. Diamond, D. M. Chen, M. L. Pritchard, B. G. Feagan, R. D. Cohen, B. A. Salzberg, and S. B. Hanauer, “Factors associated with the development of intestinal strictures or obstructions in patients with Crohn's disease,” Am. J. Gastroenterol. 101(5), 1030–1038 (2006). [CrossRef]
2. G. Oberhuber, P. C. Stangl, H. Vogelsang, E. Schober, F. Herbst, and C. Gasche, “Significant association of strictures and internal fistula formation in Crohn’s disease,” Virchows Arch. 437(3), 293–297 (2000). [CrossRef]
3. S. Kugathasan, L. A. Denson, T. D. Walters, M.-O. Kim, U. M. Marigorta, M. Schirmer, K. Mondal, C. Liu, A. Griffiths, J. D. Noe, W. V. Crandall, S. Snapper, S. Rabizadeh, J. R. Rosh, J. M. Shapiro, S. Guthery, D. R. Mack, R. Kellermayer, M. D. Kappelman, S. Steiner, D. E. Moulton, D. Keljo, S. Cohen, M. Oliva-Hemker, M. B. Heyman, A. R. Otley, S. S. Baker, J. S. Evans, B. S. Kirschner, A. S. Patel, D. Ziring, B. C. Trapnell, F. A. Sylvester, M. C. Stephens, R. N. Baldassano, J. F. Markowitz, J. Cho, R. J. Xavier, C. Huttenhower, B. J. Aronow, G. Gibson, J. S. Hyams, and M. C. Dubinsky, “Prediction of complicated disease course for children newly diagnosed with Crohn's disease: a multicentre inception cohort study,” The Lancet 389(10080), 1710–1718 (2017). [CrossRef]
4. Y. Zhu, L. A. Johnson, J. M. Rubin, H. Appelman, L. Ni, J. Yuan, X. Wang, P. D. R. Higgins, and G. Xu, “Strain-photoacoustic imaging as a potential tool for characterizing intestinal fibrosis,” Gastroenterology 157(5), 1196–1198 (2019). [CrossRef]
5. H. Lei, L. A. Johnson, K. A. Eaton, S. Liu, J. Ni, X. Wang, P. D. R. Higgins, and G. Xu, “Characterizing intestinal strictures of Crohn's disease in vivo by endoscopic photoacoustic imaging,” Biomed. Opt. Express 10(5), 2542–2555 (2019). [CrossRef]
6. H. Lei, L. A. Johnson, S. Liu, D. S. Moons, T. Ma, Q. Zhou, M. D. Rice, J. Ni, X. Wang, P. D. R. Higgins, and G. Xu, “Characterizing intestinal inflammation and fibrosis in Crohn's disease by photoacoustic imaging: feasibility study,” Biomed. Opt. Express 7(7), 2837–2848 (2016). [CrossRef]
7. H.-W. Wang, N. Chai, P. Wang, S. Hu, W. Dou, D. Umulis, L. V. Wang, M. Sturek, R. Lucht, and J.-X. Cheng, “Label-free bond-selective imaging by listening to vibrationally excited molecules,” Phys. Rev. Lett. 106(23), 238106 (2011). [CrossRef]
8. M. Wu, B. C. J. van Teeffelen, K. Ito, F. N. van de Vosse, R. P. A. Janssen, C. C. van Donkelaar, and R. G. P. Lopata, “Spectroscopic photoacoustic imaging of cartilage damage,” Osteoarthritis Cartilage 29(7), 1071–1080 (2021). [CrossRef]
9. J.-M. Yang, C. Favazza, R. Chen, J. Yao, X. Cai, K. Maslov, Q. Zhou, K. K. Shung, and L. V. Wang, “Simultaneous functional photoacoustic and ultrasonic endoscopy of internal organs in vivo,” Nat. Med. 18(8), 1297–1302 (2012). [CrossRef]
10. P. Wang, T. Ma, M. N. Slipchenko, S. Liang, J. Hui, K. K. Shung, S. Roy, M. Sturek, Q. Zhou, Z. Chen, and J.-X. Cheng, “High-speed Intravascular photoacoustic imaging of lipid-laden atherosclerotic plaque enabled by a 2-khz barium nitrite Raman laser,” Sci. Rep. 4(1), 6889 (2015). [CrossRef]
11. J.-M. Yang, C. Li, R. Chen, B. Rao, J. Yao, C.-H. Yeh, A. Danielli, K. Maslov, Q. Zhou, K. K. Shung, and L. V. Wang, “Optical-resolution photoacoustic endomicroscopy in vivo,” Biomed. Opt. Express 6(3), 918–932 (2015). [CrossRef]
12. A. Di Sabatino, R. Ciccocioppo, E. Armellini, R. Morera, L. Ricevuti, P. Cazzola, I. Fulle, and G. R. Corazza, “Serum bFGF and VEGF correlate respectively with bowel wall thickness and intramural blood flow in Crohn’s disease,” Inflammatory Bowel Diseases 10(5), 573–577 (2004). [CrossRef]
13. K. D. Bodily, J. G. Fletcher, C. A. Solem, C. D. Johnson, J. L. Fidler, J. M. Barlow, M. R. Bruesewitz, C. H. McCollough, W. J. Sandborn, E. V. Loftus, W. S. Harmsen, and B. S. Crownhart, “Crohn disease: mural attenuation and thickness at contrast-enhanced CT enterography—correlation with endoscopic and histologic findings of inflammation,” Radiology 238(2), 505–516 (2006). [CrossRef]
14. F. Hirai, A. Andoh, F. Ueno, K. Watanabe, N. Ohmiya, H. Nakase, S. Kato, M. Esaki, Y. Endo, H. Yamamoto, T. Matsui, M. Iida, T. Hibi, M. Watanabe, Y. Suzuki, and T. Matsumoto, “Efficacy of endoscopic balloon dilation for small bowel strictures in patients with Crohn’s disease: a nationwide, multi-centre, open-label, prospective cohort study,” J. Crohn's and Colitis 12(4), 394–401 (2018). [CrossRef]
15. V. V. Singh, P. Draganov, and J. Valentine, “Efficacy and safety of endoscopic balloon dilation of symptomatic upper and lower gastrointestinal Crohn's disease strictures,” J. Clin. Gastroenterol. 39(4), 284–290 (2005). [CrossRef]
16. J. Davenport, M. Hickey, J. Phillips, and P. Kyriacou, “Method for producing angled optical fiber tips in the laboratory,” Opt. Eng. 55(2), 026120 (2016). [CrossRef]
17. H. Dehghani, M. E. Eames, P. K. Yalavarthy, S. C. Davis, S. Srinivasan, C. M. Carpenter, B. W. Pogue, and K. D. Paulsen, “Near infrared optical tomography using NIRFAST: Algorithm for numerical model and image reconstruction,” Commun. Numer. Meth. Engng. 25(6), 711–732 (2009). [CrossRef]
18. D. Anthony, F. Savage, V. Sams, and P. Boulos, “The characterization of a rabbit model of inflammatory bowel disease,” Int. J. Exp. Pathol. 76(3), 215–224 (1995).
19. R. W. Stidham, J. Xu, L. A. Johnson, K. Kim, D. S. Moons, B. J. McKenna, J. M. Rubin, and P. D. Higgins, “Ultrasound elasticity imaging for detecting intestinal fibrosis and inflammation in rats and humans with Crohn's disease,” Gastroenterology 141(3), 819–826.e1 (2011). [CrossRef]
20. K. Kim, L. A. Johnson, C. Jia, J. C. Joyce, S. Rangwalla, P. D. Higgins, and J. M. Rubin, “Noninvasive ultrasound elasticity imaging (UEI) of Crohn's disease: animal model,” Ultrasound Med. Biol. 34(6), 902–912 (2008). [CrossRef]
21. J. R. Dillman, R. W. Stidham, P. D. Higgins, D. S. Moons, L. A. Johnson, and J. M. Rubin, “US elastography–derived shear wave velocity helps distinguish acutely inflamed from fibrotic bowel in a Crohn disease animal model,” Radiology 267(3), 757–766 (2013). [CrossRef]
22. Y. Zhu, L. A. Johnson, Z. Huang, J. M. Rubin, J. Yuan, H. Lei, J. Ni, X. Wang, P. D. Higgins, and G. Xu, “Identifying intestinal fibrosis and inflammation by spectroscopic photoacoustic imaging: an animal study in vivo,” Biomed. Opt. Express 9(4), 1590–1600 (2018). [CrossRef]
