Original Article
Raman Spectroscopy in the Diagnosis of Ulcerative Colitis Michelle Anne Veenstra1,2 Olena Palyvoda3 Hazem Alahwal2 Marko Jovanovski2 Luke Anthony Reisner3 Brady King1,2 Janet Poulik4 Michael D. Klein1,5 1 Department of Pediatric Surgery, Children’s Hospital of Michigan,
Detroit, Michigan, United States 2 Department of General Surgery, Wayne State University, Detroit, Michigan, United States 3 Department of Electrical and Computer Engineering, Wayne State University, Detroit, Michigan, United States 4 Department of Pathology, Children’s Hospital of Michigan, Detroit, Michigan, United States 5 Department of Surgery, Wayne State University School of Medicine, Detroit, Michigan, United States
Address for correspondence Michelle Anne Veenstra, MD, Department of Pediatric Surgery, Children’s Hospital of Michigan, 3901 Beaubien Blvd, Detroit, MI 48202, United States (e-mail: [email protected]).
Eur J Pediatr Surg 2015;25:56–59.
Abstract
Keywords
► pediatric ► inflammatory bowel disease ► ulcerative colitis ► Raman spectroscopy ► spectroscopy
received July 1, 2014 accepted after revision July 2, 2014 published online August 31, 2014
Introduction At present, the diagnosis of ulcerative colitis (UC) requires the histologic demonstration of characteristic mucosal inflammatory changes. A rapid and noninvasive diagnosis would be of value, especially if it could be adapted to a simple rectal probe. Raman spectroscopy creates a molecular fingerprint of substances by detecting laser light scattered from asymmetric, vibrating, and chemical bonds. We hypothesize that Raman spectroscopy can distinguish UC from non-UC colon tissue rapidly and accurately. Materials and Methods Colon tissue specimens were obtained from patients operated at the Children’s Hospital of Michigan, United States, including UC colon and non-UC colon. The samples were examined with a Renishaw inVia Raman microscope (Gloucestershire, United Kingdom) with a 785 nm laser. Principal component analysis and discriminant function analysis were used to classify groups. Final classification was evaluated against histologic diagnoses using leave-one-out cross-validation at a spectral level. Results We compared Raman spectroscopy examination of colon specimens from four patients with UC and four patients without UC. A total of 801 spectra were recorded from colon specimens. We evaluated 100 spectra each from the mucosal and serosal surfaces of patients with UC and 260 spectra from the mucosal surface and 341 spectra from the serosal surface of the patients who did not have UC. For samples from the mucosal surface, the Raman analysis had a sensitivity of 82% and a specificity of 89%. For samples from the serosal surface, Raman spectroscopy had a sensitivity of 87% and a specificity of 93%. When considering each tissue sample and deciding the diagnosis based on the majority of spectra from that sample, there were no errors in the diagnosis. Conclusions Raman spectroscopy can distinguish UC from normal colon tissue rapidly and accurately. This technology offers the possibility of real-time diagnosis as well as the ability to study changes in UC-afflicted colon tissue that do not appear histologically.
© 2015 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0034-1387951. ISSN 0939-7248.
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
56
Raman Spectroscopy in the Diagnosis of UC
Ulcerative colitis (UC) is an inflammatory bowel disease that is limited to the mucosa of the colon. At present, the diagnosis of UC is made by clinical data, serologic markers, imaging studies, and specific histologic changes seen in a mucosal biopsy.1,2 A technique known as Raman spectroscopy can provide detailed information about the molecular composition of the tissue.3–5 We investigated the use of Raman spectroscopy in the diagnosis of UC and in the characterization of the chemical differences between tissues with and without UC.
Materials and Methods
objective lens. A single-grating spectrograph with a 1,200 lines/mm grating combined with a holographic notch filter for Rayleigh scattering rejection was used. The spectral resolution was approximately 4 cm 1. The laser power was set at 50% (62 mW) to prevent thermal damage to the samples. Each spectrum consisted of the average of two collections with a 10-second collection time and an extended wavenumber range of 600 to 1,800 cm 1. At least 20 spectra were measured from each tissue from both the mucosal and the serosal surfaces.
Histologic Examination A pediatric pathologist, blind to the Raman spectra, reviewed all tissues to determine histologic diagnosis.
Tissue Collection
Data Processing
This study was approved by the institutional review board of Wayne State University (protocol no. 1101009224). Specimens were obtained from patients who had an operation at the Children’s Hospital of Michigan, United States, including four patients with UC and four without UC. Of the four patients without UC, two patients had a closure of colostomy created for imperforate anus and two patients had closure of colostomy created for chronic constipation, without any evidence of Hirschsprung disease. The samples for Raman analysis were stored at 80°C without fixation. An additional sample from the same tissue as the Raman sample was fixed in formalin for routine histopathology to confirm the diagnosis. The frozen specimens were thawed at room temperature in normal saline before analysis. Segments were prepared to expose the mucosa or the serosa to the microscope lens, and then placed on steel slides and mounted on the Raman microscope stage.
Raman processing software was used to import, process, and analyze the data.6 The acquired raw spectra were corrected by subtracting background fluorescence, reducing noise, and normalizing the intensities. The symmetric ring breathing mode of phenylalanine (1,000–1,003 cm 1) was present in all the measured spectra and was used as a control marker, identifying which spectra had been properly measured with correct focusing and positioning of the lens. The mean spectra were calculated and plotted for four different groups of tissue (UC mucosa, non-UC mucosa, UC serosa, and non-UC serosa). The data was analyzed using principal component analysis (PCA) and discriminant function analysis (DFA). Raman classification performance was evaluated against the histologic diagnoses using leave-one-spectrum-out cross-validation. Raman peak assignments were based on the data from several publications and reviews.4,7
Raman Spectroscopy The Raman effect is an inelastic light-scattering process. Most light photons are scattered in tissue without a change in frequency (elastic scattering), which is responsible for reflectance. However, a very small fraction of the incident light (1 in 10 million photons) is scattered inelastically such that the photon energy is changed. As the photon energy is proportional to frequency, this Raman-scattered light is shifted to a different frequency. A Raman spectrum is a plot of the intensity of Raman-scattered radiation as a function of its frequency difference from the incident radiation (expressed as a wavenumber with a unit of cm 1).4 This difference is called the Raman shift. Using a spectral analyzer, the molecular information contained in the Raman emission spectrum can be obtained from a sample. Every molecule possesses a unique pattern of Raman spectral peaks, which enables the molecular composition of a tissue sample to be assessed. More specifically, the population of common molecular bonds can be identified. Proteins, lipids, nucleic acids, and other substances all exhibit distinct Raman signatures.4 Raman measurements were acquired using an inVia Raman microscope (Renishaw, Gloucestershire, United Kingdom) with a 785 nm (infrared) excitation laser and a 50
57
Results Using Raman spectroscopy, a total of 801 spectra were recorded from specimens of colon tissues with and without UC. There were 100 spectra from the mucosal surface and 100 spectra from the serosal surface of the four patients with UC. These spectra were compared with 260 spectra from the mucosal surface and 341 spectra from the serosal surface, respectively, of the four patients who did not have UC. The spectra were analyzed using PCA, and the results, along with histopathology data, were used to analyze the groups of tissue. By using DFA and a leave-one-out analysis on the PCAgenerated data, the trained DFA classifier correctly classified 313/360 (87%) of the spectra for the mucosa and 405/441 (92%) of the spectra for the serosa. Of the UC spectra, 169/200 (85%) spectra were correctly classified and 549/601 (91%) of the non-UC tissue spectra were correctly classified. This corresponds to a sensitivity of 82% and a specificity of 89% for Raman analysis of the mucosa. For Raman analysis of the serosa, the sensitivity was 87% and the specificity was 93%. When considering each tissue sample and deciding the diagnosis based on the majority of spectra from that sample, there were no errors in diagnosis. The mean Raman spectra obtained from the mucosal and serosal surfaces of the UC and non-UC colon tissues are shown in ►Fig. 1. European Journal of Pediatric Surgery
Vol. 25
No. 1/2015
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
Introduction
Veenstra et al.
Raman Spectroscopy in the Diagnosis of UC
Veenstra et al. tion and cell turnover.8 As a result, these features could be used to predict the disease. There were changes detected by Raman spectroscopy in the serosa of UC-afflicted colons that are not detected histologically. The main spectral features in the UC serosa include bands related to proteins (1,079; 1,155; 1,256; and 1,386 cm 1), lipids (1,079; 1,314; and 1,331 cm 1), and nucleic acids (896; 1,079; 1,098; 1,256; and 1,331 cm 1). The normal serosa has distinguishable peaks at 610 cm 1 (cholesterol), 1,081 cm 1 (phospholipids, collagen, and nucleic acids), 1,262 cm 1 (proteins and lipids), and 1,306 cm 1 (proteins). These features could be considered as independent predictors of disease activity in patients with UC and may represent a link between inflammation and UC.
Conclusion Many aspects of UC and other inflammatory bowel diseases such as Crohn disease present clinical challenges.9–11 There are no “gold standard” tests or examinations for the diagnosis, prognosis, assessment of disease severity, or evaluation of the effects of therapy.12–16 Our data suggest that Raman spectroscopy will be able to play an adjunctive or primary role in all of these areas and can be reliably used for diagnosis of UC.
Conflict of Interest None.
Fig. 1 Mean Raman spectra of the mucosal (a) and serosal (b) surfaces of colon tissues with and without ulcerative colitis.
References 1 Danese S, Fiocchi C. Ulcerative colitis. N Engl J Med 2011;365(18):
1713–1725 2 Gisbert JP, González-Lama Y, Maté J. Role of biological markers in
Discussion Significant differences were observed between the mucosal spectra of tissues with and without UC in the spectral regions associated with nucleic acids, proteins, and lipids. In the serosa, there were fewer differences between the UC and normal spectra, most likely because UC primarily affects the mucosa.1 The main spectral differences between diseased and normal mucosa samples were in the Raman shift regions of proteins (755; 1,030–1,032; 1,102; 1,170–1,173; 1,246; 1,613; and 1,656–1,658 cm 1), lipids (1,102 and 1,656– 1,658 cm- 1), and nucleic acids (1,044; 1,102; and 1,170– 1,173 cm 1).4,7 Significant peaks were observed that are only present in UC-afflicted mucosal colonic tissues and can be assigned to proteins (755; 1,102; 1,246; and 1,613 cm 1), lipids (1,102 and 1,246 cm 1), and nucleic acids (1,044 cm 1).4,7 The normal mucosa showed peaks at 1,308 and 1,741 cm 1, corresponding to C–N asymmetric stretching in asymmetric aromatic amines and ester groups from fatty acids, respectively.4,7 The Raman spectra indicate that the intensities of protein and nucleic acids bands are increased in mucosal UC tissues, which could be explained by inflamma-
European Journal of Pediatric Surgery
Vol. 25
No. 1/2015
3
4
5
6
7
8
9 10 11
inflammatory bowel disease [in Spanish]. Gastroenterol Hepatol 2007;30(3):117–129 Rodriguez LG, Lockett SJ, Holtom GR. Coherent anti-stokes Raman scattering microscopy: a biological review. Cytometry A 2006; 69(8):779–791 Movasaghi Z, Rehman S, Rehman IU. Raman spectroscopy of biological tissues. Appl Spectrosc Rev 2007;42(5): 493–541 Lawson EE, Barry BW, Williams AC, et al. Biomedical applications of Raman spectroscopy. J Raman Spectrosc 1997;28(2–3): 111–117 Reisner LA, Cao A, Pandya AK. An integrated software system for processing, analyzing, and classifying Raman spectra. Chemom Intell Lab Syst 2011;105(1):83–90 Socrates G. Infrared and Raman Characteristic Group Frequencies: Tables and Charts. West Sussex, England: John Wiley & Sons Ltd; 2004 Zhang SZ, Zhao XH, Zhang DC. Cellular and molecular immunopathogenesis of ulcerative colitis. Cell Mol Immunol 2006;3(1): 35–40 Hanauer SB. Inflammatory bowel disease. N Engl J Med 1996; 334(13):841–848 Gibson PR, Muir JG. Reinforcing the mucus: a new therapeutic approach for ulcerative colitis? Gut 2005;54(7):900–903 Mehta SJ, Silver AR, Lindsay JO. Review article: strategies for the management of chronic unremitting ulcerative colitis. Aliment Pharmacol Ther 2013;38(2):77–97
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
58
Raman Spectroscopy in the Diagnosis of UC
histochemical markers in predicting progression of dysplasia to advanced neoplasia in patients with ulcerative colitis. Inflamm Bowel Dis 2012;18(3):480–488 13 Yun J, Xu CT, Pan BR. Epidemiology and gene markers of ulcerative colitis in the Chinese. World J Gastroenterol 2009; 15(7):788–803
59
14 Roy HK, Backman V. Spectroscopic applications in gastrointestinal
endoscopy. Clin Gastroenterol Hepatol 2012;10(12):1335–1341 15 Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and
cancer. Cell 2010;140(6):883–899 16 Kiesslich R, Neurath MF. Endoscopic detection of early lower
gastrointestinal cancer. Best Pract Res Clin Gastroenterol 2005; 19(6):941–961
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
12 van Schaik FD, Oldenburg B, Offerhaus GJ, et al. Role of immuno-
Veenstra et al.
European Journal of Pediatric Surgery
Vol. 25
No. 1/2015
Raman Spectroscopy in the Diagnosis of Ulcerative Colitis Michelle Anne Veenstra1,2 Olena Palyvoda3 Hazem Alahwal2 Marko Jovanovski2 Luke Anthony Reisner3 Brady King1,2 Janet Poulik4 Michael D. Klein1,5 1 Department of Pediatric Surgery, Children’s Hospital of Michigan,
Detroit, Michigan, United States 2 Department of General Surgery, Wayne State University, Detroit, Michigan, United States 3 Department of Electrical and Computer Engineering, Wayne State University, Detroit, Michigan, United States 4 Department of Pathology, Children’s Hospital of Michigan, Detroit, Michigan, United States 5 Department of Surgery, Wayne State University School of Medicine, Detroit, Michigan, United States
Address for correspondence Michelle Anne Veenstra, MD, Department of Pediatric Surgery, Children’s Hospital of Michigan, 3901 Beaubien Blvd, Detroit, MI 48202, United States (e-mail: [email protected]).
Eur J Pediatr Surg 2015;25:56–59.
Abstract
Keywords
► pediatric ► inflammatory bowel disease ► ulcerative colitis ► Raman spectroscopy ► spectroscopy
received July 1, 2014 accepted after revision July 2, 2014 published online August 31, 2014
Introduction At present, the diagnosis of ulcerative colitis (UC) requires the histologic demonstration of characteristic mucosal inflammatory changes. A rapid and noninvasive diagnosis would be of value, especially if it could be adapted to a simple rectal probe. Raman spectroscopy creates a molecular fingerprint of substances by detecting laser light scattered from asymmetric, vibrating, and chemical bonds. We hypothesize that Raman spectroscopy can distinguish UC from non-UC colon tissue rapidly and accurately. Materials and Methods Colon tissue specimens were obtained from patients operated at the Children’s Hospital of Michigan, United States, including UC colon and non-UC colon. The samples were examined with a Renishaw inVia Raman microscope (Gloucestershire, United Kingdom) with a 785 nm laser. Principal component analysis and discriminant function analysis were used to classify groups. Final classification was evaluated against histologic diagnoses using leave-one-out cross-validation at a spectral level. Results We compared Raman spectroscopy examination of colon specimens from four patients with UC and four patients without UC. A total of 801 spectra were recorded from colon specimens. We evaluated 100 spectra each from the mucosal and serosal surfaces of patients with UC and 260 spectra from the mucosal surface and 341 spectra from the serosal surface of the patients who did not have UC. For samples from the mucosal surface, the Raman analysis had a sensitivity of 82% and a specificity of 89%. For samples from the serosal surface, Raman spectroscopy had a sensitivity of 87% and a specificity of 93%. When considering each tissue sample and deciding the diagnosis based on the majority of spectra from that sample, there were no errors in the diagnosis. Conclusions Raman spectroscopy can distinguish UC from normal colon tissue rapidly and accurately. This technology offers the possibility of real-time diagnosis as well as the ability to study changes in UC-afflicted colon tissue that do not appear histologically.
© 2015 Georg Thieme Verlag KG Stuttgart · New York
DOI http://dx.doi.org/ 10.1055/s-0034-1387951. ISSN 0939-7248.
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
56
Raman Spectroscopy in the Diagnosis of UC
Ulcerative colitis (UC) is an inflammatory bowel disease that is limited to the mucosa of the colon. At present, the diagnosis of UC is made by clinical data, serologic markers, imaging studies, and specific histologic changes seen in a mucosal biopsy.1,2 A technique known as Raman spectroscopy can provide detailed information about the molecular composition of the tissue.3–5 We investigated the use of Raman spectroscopy in the diagnosis of UC and in the characterization of the chemical differences between tissues with and without UC.
Materials and Methods
objective lens. A single-grating spectrograph with a 1,200 lines/mm grating combined with a holographic notch filter for Rayleigh scattering rejection was used. The spectral resolution was approximately 4 cm 1. The laser power was set at 50% (62 mW) to prevent thermal damage to the samples. Each spectrum consisted of the average of two collections with a 10-second collection time and an extended wavenumber range of 600 to 1,800 cm 1. At least 20 spectra were measured from each tissue from both the mucosal and the serosal surfaces.
Histologic Examination A pediatric pathologist, blind to the Raman spectra, reviewed all tissues to determine histologic diagnosis.
Tissue Collection
Data Processing
This study was approved by the institutional review board of Wayne State University (protocol no. 1101009224). Specimens were obtained from patients who had an operation at the Children’s Hospital of Michigan, United States, including four patients with UC and four without UC. Of the four patients without UC, two patients had a closure of colostomy created for imperforate anus and two patients had closure of colostomy created for chronic constipation, without any evidence of Hirschsprung disease. The samples for Raman analysis were stored at 80°C without fixation. An additional sample from the same tissue as the Raman sample was fixed in formalin for routine histopathology to confirm the diagnosis. The frozen specimens were thawed at room temperature in normal saline before analysis. Segments were prepared to expose the mucosa or the serosa to the microscope lens, and then placed on steel slides and mounted on the Raman microscope stage.
Raman processing software was used to import, process, and analyze the data.6 The acquired raw spectra were corrected by subtracting background fluorescence, reducing noise, and normalizing the intensities. The symmetric ring breathing mode of phenylalanine (1,000–1,003 cm 1) was present in all the measured spectra and was used as a control marker, identifying which spectra had been properly measured with correct focusing and positioning of the lens. The mean spectra were calculated and plotted for four different groups of tissue (UC mucosa, non-UC mucosa, UC serosa, and non-UC serosa). The data was analyzed using principal component analysis (PCA) and discriminant function analysis (DFA). Raman classification performance was evaluated against the histologic diagnoses using leave-one-spectrum-out cross-validation. Raman peak assignments were based on the data from several publications and reviews.4,7
Raman Spectroscopy The Raman effect is an inelastic light-scattering process. Most light photons are scattered in tissue without a change in frequency (elastic scattering), which is responsible for reflectance. However, a very small fraction of the incident light (1 in 10 million photons) is scattered inelastically such that the photon energy is changed. As the photon energy is proportional to frequency, this Raman-scattered light is shifted to a different frequency. A Raman spectrum is a plot of the intensity of Raman-scattered radiation as a function of its frequency difference from the incident radiation (expressed as a wavenumber with a unit of cm 1).4 This difference is called the Raman shift. Using a spectral analyzer, the molecular information contained in the Raman emission spectrum can be obtained from a sample. Every molecule possesses a unique pattern of Raman spectral peaks, which enables the molecular composition of a tissue sample to be assessed. More specifically, the population of common molecular bonds can be identified. Proteins, lipids, nucleic acids, and other substances all exhibit distinct Raman signatures.4 Raman measurements were acquired using an inVia Raman microscope (Renishaw, Gloucestershire, United Kingdom) with a 785 nm (infrared) excitation laser and a 50
57
Results Using Raman spectroscopy, a total of 801 spectra were recorded from specimens of colon tissues with and without UC. There were 100 spectra from the mucosal surface and 100 spectra from the serosal surface of the four patients with UC. These spectra were compared with 260 spectra from the mucosal surface and 341 spectra from the serosal surface, respectively, of the four patients who did not have UC. The spectra were analyzed using PCA, and the results, along with histopathology data, were used to analyze the groups of tissue. By using DFA and a leave-one-out analysis on the PCAgenerated data, the trained DFA classifier correctly classified 313/360 (87%) of the spectra for the mucosa and 405/441 (92%) of the spectra for the serosa. Of the UC spectra, 169/200 (85%) spectra were correctly classified and 549/601 (91%) of the non-UC tissue spectra were correctly classified. This corresponds to a sensitivity of 82% and a specificity of 89% for Raman analysis of the mucosa. For Raman analysis of the serosa, the sensitivity was 87% and the specificity was 93%. When considering each tissue sample and deciding the diagnosis based on the majority of spectra from that sample, there were no errors in diagnosis. The mean Raman spectra obtained from the mucosal and serosal surfaces of the UC and non-UC colon tissues are shown in ►Fig. 1. European Journal of Pediatric Surgery
Vol. 25
No. 1/2015
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
Introduction
Veenstra et al.
Raman Spectroscopy in the Diagnosis of UC
Veenstra et al. tion and cell turnover.8 As a result, these features could be used to predict the disease. There were changes detected by Raman spectroscopy in the serosa of UC-afflicted colons that are not detected histologically. The main spectral features in the UC serosa include bands related to proteins (1,079; 1,155; 1,256; and 1,386 cm 1), lipids (1,079; 1,314; and 1,331 cm 1), and nucleic acids (896; 1,079; 1,098; 1,256; and 1,331 cm 1). The normal serosa has distinguishable peaks at 610 cm 1 (cholesterol), 1,081 cm 1 (phospholipids, collagen, and nucleic acids), 1,262 cm 1 (proteins and lipids), and 1,306 cm 1 (proteins). These features could be considered as independent predictors of disease activity in patients with UC and may represent a link between inflammation and UC.
Conclusion Many aspects of UC and other inflammatory bowel diseases such as Crohn disease present clinical challenges.9–11 There are no “gold standard” tests or examinations for the diagnosis, prognosis, assessment of disease severity, or evaluation of the effects of therapy.12–16 Our data suggest that Raman spectroscopy will be able to play an adjunctive or primary role in all of these areas and can be reliably used for diagnosis of UC.
Conflict of Interest None.
Fig. 1 Mean Raman spectra of the mucosal (a) and serosal (b) surfaces of colon tissues with and without ulcerative colitis.
References 1 Danese S, Fiocchi C. Ulcerative colitis. N Engl J Med 2011;365(18):
1713–1725 2 Gisbert JP, González-Lama Y, Maté J. Role of biological markers in
Discussion Significant differences were observed between the mucosal spectra of tissues with and without UC in the spectral regions associated with nucleic acids, proteins, and lipids. In the serosa, there were fewer differences between the UC and normal spectra, most likely because UC primarily affects the mucosa.1 The main spectral differences between diseased and normal mucosa samples were in the Raman shift regions of proteins (755; 1,030–1,032; 1,102; 1,170–1,173; 1,246; 1,613; and 1,656–1,658 cm 1), lipids (1,102 and 1,656– 1,658 cm- 1), and nucleic acids (1,044; 1,102; and 1,170– 1,173 cm 1).4,7 Significant peaks were observed that are only present in UC-afflicted mucosal colonic tissues and can be assigned to proteins (755; 1,102; 1,246; and 1,613 cm 1), lipids (1,102 and 1,246 cm 1), and nucleic acids (1,044 cm 1).4,7 The normal mucosa showed peaks at 1,308 and 1,741 cm 1, corresponding to C–N asymmetric stretching in asymmetric aromatic amines and ester groups from fatty acids, respectively.4,7 The Raman spectra indicate that the intensities of protein and nucleic acids bands are increased in mucosal UC tissues, which could be explained by inflamma-
European Journal of Pediatric Surgery
Vol. 25
No. 1/2015
3
4
5
6
7
8
9 10 11
inflammatory bowel disease [in Spanish]. Gastroenterol Hepatol 2007;30(3):117–129 Rodriguez LG, Lockett SJ, Holtom GR. Coherent anti-stokes Raman scattering microscopy: a biological review. Cytometry A 2006; 69(8):779–791 Movasaghi Z, Rehman S, Rehman IU. Raman spectroscopy of biological tissues. Appl Spectrosc Rev 2007;42(5): 493–541 Lawson EE, Barry BW, Williams AC, et al. Biomedical applications of Raman spectroscopy. J Raman Spectrosc 1997;28(2–3): 111–117 Reisner LA, Cao A, Pandya AK. An integrated software system for processing, analyzing, and classifying Raman spectra. Chemom Intell Lab Syst 2011;105(1):83–90 Socrates G. Infrared and Raman Characteristic Group Frequencies: Tables and Charts. West Sussex, England: John Wiley & Sons Ltd; 2004 Zhang SZ, Zhao XH, Zhang DC. Cellular and molecular immunopathogenesis of ulcerative colitis. Cell Mol Immunol 2006;3(1): 35–40 Hanauer SB. Inflammatory bowel disease. N Engl J Med 1996; 334(13):841–848 Gibson PR, Muir JG. Reinforcing the mucus: a new therapeutic approach for ulcerative colitis? Gut 2005;54(7):900–903 Mehta SJ, Silver AR, Lindsay JO. Review article: strategies for the management of chronic unremitting ulcerative colitis. Aliment Pharmacol Ther 2013;38(2):77–97
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
58
Raman Spectroscopy in the Diagnosis of UC
histochemical markers in predicting progression of dysplasia to advanced neoplasia in patients with ulcerative colitis. Inflamm Bowel Dis 2012;18(3):480–488 13 Yun J, Xu CT, Pan BR. Epidemiology and gene markers of ulcerative colitis in the Chinese. World J Gastroenterol 2009; 15(7):788–803
59
14 Roy HK, Backman V. Spectroscopic applications in gastrointestinal
endoscopy. Clin Gastroenterol Hepatol 2012;10(12):1335–1341 15 Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and
cancer. Cell 2010;140(6):883–899 16 Kiesslich R, Neurath MF. Endoscopic detection of early lower
gastrointestinal cancer. Best Pract Res Clin Gastroenterol 2005; 19(6):941–961
Downloaded by: Wegner Health Science Information Center. Copyrighted material.
12 van Schaik FD, Oldenburg B, Offerhaus GJ, et al. Role of immuno-
Veenstra et al.
European Journal of Pediatric Surgery
Vol. 25
No. 1/2015
