OSA's Digital Library

Optics Express

Optics Express

  • Editor: Michael Duncan
  • Vol. 14, Iss. 6 — Mar. 20, 2006
  • pp: 2211–2219
« Show journal navigation

Multimodal near infrared spectral imaging as an exploratory tool for dysplastic esophageal lesion identification

Chad A. Lieber, Shiro Urayama, Nazir Rahim, Raymond Tu, Ramez Saroufeem, Boris Reubner, and Stavros G. Demos  »View Author Affiliations


Optics Express, Vol. 14, Issue 6, pp. 2211-2219 (2006)
http://dx.doi.org/10.1364/OE.14.002211


View Full Text Article

Acrobat PDF (1387 KB)





Browse Journals / Lookup Meetings

Browse by Journal and Year


   


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools

Share
Citations

Abstract

We explore nine different combinations of fluorescence, light scattering, and polarization spectral imaging approaches in the near-infrared spectral region toward the diagnosis of pathologic and normal esophageal lesions. The combinations of all the imaging techniques were evaluated for maximal sensitivity and specificity. The results suggest that this multimodal approach is capable of highly accurate detection of the presence of pathologic tissue.

© 2006 Optical Society of America

1. Introduction

Cancers of the esophagus present an estimated 14,520 new cases and account for more than 13,570 deaths annually [1

1. American Cancer Society. Cancer Facts and Figures 2005, Atlanta: American Cancer Society (2005).

]. As with most cancers, early diagnosis of esophageal malignancies is critical to a favorable prognosis. Random surveillance biopsies are the current gold standard for the identification of esophageal lesions in pre-neoplastic conditions such as Barrett’s esophagus. This method is minimally invasive; however, it is prone to sampling error, time-consuming, and cost-inefficient. A diagnostic tool that could provide rapid, automated classification of esophageal lesions would increase the patient comfort, efficiency and comprehensiveness of malignancy screening and surveillance procedures.

A variety of optical techniques have recently been utilized for tissue diagnostics. Fluorescence spectroscopy [2–4

2. M. Panjehpour, B. Overholt, T. Vo-Dinh, R. Haggitt, D. Edwards, and F. Buckley, “Endoscopic fluorescence detection of high-grade dysplasia in Barrett’s esophagus,” Gastroenterology 111, 93–101 (1996). [CrossRef] [PubMed]

], Raman spectroscopy [5

5. M. G. Shim, L. M. Song, N. E. Marcon, and B. C. Wilson, “In vivo near-infrared Raman spectroscopy: demonstration of feasibility during clinical gastrointestinal endoscopy,” Photochem. Photobiol. 72, 146–150 (2000). [PubMed]

, 6

6. C. Kendall, N. Stone, N. Shepherd, K. Geboes, B. Warren, R. Bennett, and H. Barr, “Raman spectroscopy, a potential tool for the objective identification and classification of neoplasia in Barrett’s oesophagus,” J. Pathol. 200, 602–609 (2003). [CrossRef] [PubMed]

], light scattering spectroscopy [4

4. I. Georgakoudi, B. C. Jacobson, J. Van Dam, V. Backman, M. B. Wallace, M. G. Muller, Q. Zhang, K. Badizadegan, D. Sun, G. A. Thomas, L. T. Perelman, and M. S. Feld, “Fluorescence, reflectance, and light-scattering spectroscopy for evaluating dysplasia in patients with Barrett’s esophagus,” Gastroenterology 120, 1620–1629 (2001). [CrossRef] [PubMed]

, 7

7. M. B. Wallace, L. T. Perelman, V. Backman, J. M. Crawford, M. Fitzmaurice, M. Seiler, K. Badizadegan, S. J. Shields, I. Itzkan, R. R. Dasari, J. Van Dam, and M. S. Feld, “Endoscopic detection of dysplasia in patients with Barrett’s esophagus using light-scattering spectroscopy,” Gastroenterology 119, 677–682 (2000). [CrossRef] [PubMed]

, 8

8. D. C. Pickard, I. J. Bigio, S. G. Bown, L. B. Lovat, P. M. Ripley, and M. Novelli, “Optical biopsy for the diagnosis of dysplasia in Barrett’s oesophagus,” in Optical Biopsy IV ,R. R. Alfono ed., Proc. SPIE 4613, 234–243, (2002). [CrossRef]

], and Fourier-transform infrared spectroscopy [9

9. J. S. Wang, J. S. Shi, Y. Z. Xu, X. Y Duan, L. Zhang, J. Wang, L. M. Yang, S. F. Weng, and J. G. Wu, “FT-IR spectroscopic analysis of normal and cancerous tissues of esophagus,” World J. Gastroenterol. 9, 1897–1899 (2003). [PubMed]

, 10

10. Q. B. Li, X. J. Sun, Y. Z. Xu, L. M. Yang, Y. F. Zhang, S. F. Weng, J. S. Shi, and J. G. Wu, “Diagnosis of gastric inflammation and malignancy in Endoscopic biopsies based on Fourier transform infrared Spectroscopy,” Clin. Chem. 51, 346–350 (2005). [CrossRef] [PubMed]

] are some of the techniques that have been utilized for the study of esophageal lesions. Optical spectroscopies such as these are capable of providing biochemical and morphological information in very short integration times, which can be used for automated diagnosis of intact tissue. The successes of these optical studies vary, but all have shown tremendous promise towards development of a minimally invasive, efficient diagnostic device. To be useful as a comprehensive screening device, however, the optical technique must allow rapid imaging of a large area of tissue such that suspicious regions could be identified and biopsied for positive histopathologic correlation.

In this study, we compared the efficiency of a number of near-infrared (NIR) spectral imaging approaches toward the diagnosis of pathologic and normal esophageal lesions. The experiments were carried out using a portable imaging system to explore nine different combinations of fluorescence, light scattering, and polarization modalities to image fresh biopsies of esophagus tissue. The results of this study suggest that this multimodal approach is capable of highly accurate detection of regions of pathologic tissue.

2. Experimental methods

Tissue samples were obtained from patients undergoing esophageal endoscopy at the University of California at Davis Medical Center (UCDMC). Twenty patients were included in this study. For each patient, two sets of biopsies, one from visibly normal mucosa and another from suspected abnormal tissue, were obtained. Thus, a total of 40 sample sets were collected, with each sample set containing at least 3 biopsies. All tissue sample sets were obtained immediately after biopsy and stored in sterile saline solution at ambient temperature prior to spectral imaging (<10 min.). The samples were arranged on a glass slide in groups of suspected normal and abnormal samples. A second glass slide was applied on top of these samples to hold and compress them to a uniform thickness (~1 mm) during the imaging procedures. Upon completion of the optical measurements, the samples were placed in a standard buffered formalin solution for histopathologic analysis. Pathological diagnosis of each set was determined and confirmed by at least two experienced gastrointestinal pathologists and taken as the gold standard. The 40 sample sets were comprised of 18 normal, 5 Barrett’s esophagus, 3 other benign lesions, 1 low-grade dysplasia (LGD), 2 high-grade dysplasia (HGD), 1 squamous cell carcinoma, and 10 adenocarcinoma. All protocols were approved by the UCDMC Institutional Review Board.

A portable spectral imaging system was utilized to allow rapid imaging of the esophageal samples immediately after biopsy. The system, as shown in Fig. 1, allowed imaging at multiple excitation wavelengths and polarizations. It also contained portable, low power (≈ 5 mW output) laser sources at 408 (diode), 532 (diode), and 633 nm (HeNe), as well as a broadband white light source equipped with a filter wheel. For this study, the filter wheel contained bandpass filters centered at 700, 850, and 1000 nm, each with 40 nm bandwidth. The bandpass-filtered white light source was linearly polarized by plate polarizers placed in the illumination path. All the light sources were expanded via positive lenses to illuminate the entirety of the slide holder in which the samples were placed. Images were collected by a camera lens attached to a liquid nitrogen-cooled CCD detector. A longpass filter with cut-on wavelength of 700 nm was placed in the collection leg to assure collection of near-infrared emission, while a removable analyzing polarizer allowed selection of parallel or orthogonal polarization components.

For the acquisition of the NIR autofluorescence measurements, each set of specimens were illuminated with one of the three laser sources (408, 532, or 633 nm) and an image was acquired using 40 second integration time; this was repeated for each laser source. The light scattering measurements were made using the broadband light source after passing through each of the three bandpass filters (700, 850, or 1000 nm). The polarization orientation of the analyzer was positioned parallel to the polarization angle of the illuminator. Images were acquired using 0.2-second integration times, and repeated for each illumination wavelength. The polarization orientation of the analyzer was then rotated orthogonal to the polarization angle of the illuminator and images were acquired using 0.2-second integration times for each of the three illumination wavelengths. Thus a total of nine images were obtained of each sample set: three fluorescence images, three parallel polarization light scattering images, and three orthogonal polarization light scattering images. Additionally, three degree-of-polarization (DOP) images were created for each sample set by nonlinear mathematical combination of the polarization images as the ratio of the polarization difference (parallel minus orthogonal) to the polarization sum (parallel plus orthogonal) for each illumination wavelength. All images were stored in a computer database for subsequent analysis.

Each patient’s set of twelve images was processed independently to delineate the margins of the biopsy samples within each image. The image intensity was first normalized using a reference image of the illumination intensity for each imaging method to account for nonuniformity of the illumination. One image from each patient was loaded and displayed, and feature masks were created via a graphical user interface created specifically for this experiment using the MATLAB software application. Two masks were created for each image set: margins of suspected normal biopsies and margins of suspected abnormal biopsies. Each of the two masks was then applied to each of the twelve images for the respective patient, such that the mean intensity within each mask was calculated for each image. This reduced each image to two normalized intensities: normalized intensity of the presumed normal sample, and normalized intensity of the presumed abnormal sample. This process was repeated for each of the 20 image sets (40 sample sets).

Fig. 1. Schematic of imaging system. Illumination is provided by compact laser sources at 408, 532, and 633 nm and a broadband light source that is wavelength-filtered by a filter wheel (FW) containing bandpass filters at 700, 850, and 1000 nm before being transmitted through a fiber bundle (FB) and polarizer. The collected light passes through an analyzing and a 700 nm longpass filter (LP) before being imaged onto a cooled CCD detector via a camera lens (CL).

The 40 sample sets were divided into two groups: “low-risk” containing normal, Barrett’s mucosa without dysplasia, and other benign samples; “high-risk” containing LGD, HGD, squamous cell carcinoma, and adenocarcinoma cases. The normalized intensities for each sample set were used as inputs into a discriminant diagnostic algorithm for classification. A logistic discriminant function was utilized, which regressed a binomial distribution curve on the samples to maximally separate the two classes. Cross-validation was achieved using a leave-one-out procedure. All possible combinations of the twelve images were evaluated for maximum classification accuracy via several iterations of an n-choose-k algorithm, with n=12 and k=2 through 12.

3. Experimental results

The normalized tissue intensity values from each imaging modality were plotted to ascertain whether any of the imaging techniques could be used independently to separate the low-risk from the high-risk samples. The normalized intensities of the samples for each of the nine image modalities and the three DOP combinations were as shown in Fig. 4. In the raw univariate analysis, the trend of higher intensities among the high-risk samples versus the low-risk samples was again evident in all of the imaging modalities. However, no single imaging technique was able to provide a clear distinction between the tissue sets. Thus a multivariate analysis was employed to allow mathematical combinations of the images for classification.

Fig. 2. Representative images of esophageal biopsy specimens acquired via nine different imaging modalities. Fluorescence images were acquired using excitation at (A) 408 nm, (B) 532 nm, and (C) 633 nm. Light scattering images were acquired using illumination at 700 nm via (D) parallel- and (E) cross-polarized scattering, 850 nm via (F) parallel- and (G) cross-polarized scattering, and 1000 nm via (H) parallel- and (I) cross-polarized scattering. Biopsy set on the left half of these specific images is pathologically normal (sample #15), and the set on the right is adenocarcinoma (sample #35).
Fig. 3. Representative degree-of-polarization (DOP) images mathematically constructed from the polarization difference to polarization sum ratios at (A) 700 nm, (B) 850 nm, and (C) 1000 nm shown in Fig. 2.

Every possible combination of the twelve normalized intensities (4083 total) from each sample set was input into a logistic discriminant analysis for classification, using leave-one-out cross-validation. The image combinations were evaluated first for maximum sensitivity towards detection of the high-risk samples, and resulted in several combinations correctly classifying 12 of the 14 high-risk samples. These combinations with maximal sensitivity were then similarly compared for specificity, and resulted in a single combination of five images that classified 25 of the 26 low-risk samples correctly. The five images in this combination included the 1000 nm parallel- and cross-polarized, 633 nm fluorescence, 850 nm cross-polarized, and 700 nm DOP images. Table 1 shows the pathologies of the 40 sample sets studied and their a priori assigned risk categories, along with the predicted classifications as determined by the logistic discriminant function using the selected five images as input. This discriminant function achieved a sensitivity of 86%, and a specificity of 96%. The logistic discriminant scores assigned to each sample set were as shown in Fig. 5. This analysis classified low-risk samples as those with scores less than 0.5, while high-risk samples were those with scores greater than 0.5. The misclassified normal sample was assigned a score of 0.89, while the misclassified adenocarcinoma was 1.6×10-7 and the HGD was 6.4×10-5.

Fig. 4. Plot of raw image intensities from each biopsy set as a function of imaging modality. Intensities were normalized to maximum and minimum for each respective imaging modality. Labels correspond to excitation/illumination wavelength (in nm) and imaging technique: (F) fluorescence, (∥) parallel-polarized light scattering, (+) cross-polarized light scattering, and (D) degree-of-polarization. A general trend of higher intensities among the high-risk samples is evident in most of the imaging modalities, but none of the individual modalities provide a clear differentiation of the two risk classes.
Fig. 5. Plot of output scores from the logistic discriminant algorithm for all of the 40 biopsy sets studied. Scores correspond to the probability of high-risk classification, with those scoring above 0.5 classified as high risk and those scoring below 0.5 classified as low risk. Arrows indicate the three misclassified samples: one pathologically normal sample scored at 0.89, one adenocarcinoma and one high-grade dysplasia scored very near zero.

4. Discussion

The multimodal spectral imaging system utilized for this study allowed nine different imaging techniques to be used in quick succession for the acquisition of spectral images of the esophageal specimens. Via electronically controlled shutters and filter wheels the nine images were acquired in less than two minutes. However, the image acquisition time can be reduced to just a few seconds if we use higher power laser sources (e.g. 100 mW compared to current 5 mW output). The portability of the system allowed it to be placed outside the examination room so the tissue specimens could be imaged immediately after acquiring biopsies. This timing is critical, as the autofluorescence intensity of tissue is known to change after removal from the body. The rapid imaging also allowed a direct comparison of the individual imaging modalities, as well as mathematical combination of the spectral images.

The spectral images showed a general increase in intensity from the high-risk esophageal lesions as compared to the low-risk benign and normal samples. But while this general trend of increased intensity amongst the high-risk tissues was present, no single imaging modalities used in this study provided a consistently clear distinction between the low- and high-risk samples. This finding was verified using a simple univariate analysis of the normalized image intensities from each imaging modality. However, by imaging the tissues in precisely the same position and orientation during each of the nine acquisitions, multivariate analysis was successful in classifying malignant and benign tissues to take advantage of the biochemical and morphological differences as probed by each imaging modality.

A variety of physiological and morphological transformations are known to occur in the transition from normal to dysplasia to malignancy, and can be theorized to account for the observed differences in the scattering intensity. Increased nuclear size and proliferation of cells would account for such an increase in scattering intensity in the high-risk samples versus the low-risk samples in this study [11

11. L. T. Perelman, V. Backman, M. Wallace, G. Zonios, R. Manoharan, A. Nusrat, S. Shields, M. Seiler, C. Lima, T. Hamano, I. Itzkan, J. Van Dam, J. M. Crawford, and M. S. Feld, “Observation of periodic fine structure in reflectance from biological tissue: A new technique for measuring nuclear size distribution,” Phys. Rev. Lett. 80, 627–630 (1998). [CrossRef]

, 12

12. H. Fang, M. Ollero, E. Vitkin, L. M. Kimerer, P. B. Cipolloni, M. M. Zaman, S. D. Freedman, I. J. Bigio, I. Itzkan, E. B. Hanlon, and L. T. Perelman, “Noninvasive sizing of subcellular organelles with light scattering spectroscopy,” IEEE J. Sel. Top. Quantum Electron. 9, 267–276 (2003). [CrossRef]

]. The proliferation of cells not only increases the overall nuclear mass of the interrogated region, but also corresponds to an increase in variation in nuclear shape (pleomorphism) and chromatin concentration (hyperchromasia), both of which increase scattering intensity [13

13. R. S. Gurjar, V. Backman, L. T. Perelman, I. Georgakoudi, K. Badizadegan, I. Itzkan, R. R. Dasari, and M. S. Feld, “Imaging human epithelial properties with polarized light-scattering spectroscopy,” Nat. Med. 7, 1245–1248 (2001). [CrossRef] [PubMed]

].

All tissue specimens used in this work were thin sections and their thickness after they were positioned between the glass slides in the sample holder was ~1 mm or less. As a result, while the thickness of all specimens of each set was the same, the thickness between sets obtained from different patients varied. Given that the images were formed by photons emitted in the NIR spectral region, which has a tissue penetration depth much greater than the ~1 mm specimen thickness, it is expected that the thickness of each specimen affected the measured average intensity for all imaging methods utilized in this work. In our analysis of the experimental data we did not take this aspect into account because such a variation is typical of that found in a clinical setting. However, our experimental results suggest that this multi-modal optical interrogation approach can compensate for the variability of the lesion thickness, as evidenced by the high classification accuracy.

Contrarily, the large field of view of the imaging system used in this study could also be a reason for the classification errors. Final pathological diagnosis of the lesions was determined as the most advanced pathology grade evident in the entire field of biopsies in each set. In the data processing phase of this study, each set of biopsies was reduced to single mean image intensity for each imaging modality. Therefore, the field of view was mathematically determined as the entirety of each set of biopsies. Thus, it is possible that the misclassified HGD and adenocarcinoma were focal lesions too small to affect the mean image intensity over all of the biopsies with which they were contained. In future work, we will take full advantage of the benefit of using an imaging method by acquiring images of each biopsy sample independently, with histopathologic correlation of each sample and extend the image resolution to the microscopic level. As this was a proof-of-concept experiment, such correlation was not available for this study.

The benefits of a multimodal approach for classification of esophageal lesions have been demonstrated by Georgakoudi et al. [4

4. I. Georgakoudi, B. C. Jacobson, J. Van Dam, V. Backman, M. B. Wallace, M. G. Muller, Q. Zhang, K. Badizadegan, D. Sun, G. A. Thomas, L. T. Perelman, and M. S. Feld, “Fluorescence, reflectance, and light-scattering spectroscopy for evaluating dysplasia in patients with Barrett’s esophagus,” Gastroenterology 120, 1620–1629 (2001). [CrossRef] [PubMed]

]. Using a combination of fluorescence and light scattering spectroscopies in point-based measurements, they were able to classify high- and low-grade dysplasias with much higher accuracy than with the individual spectroscopic techniques. In particular, that study utilized differences in fluorescence in the red-visible and blue-green wavelengths as well as changes in reduced scattering coefficient. These benefits were mirrored in this imaging study, where five imaging techniques classified 37 out of 40 samples correctly. However, this work extends beyond the work by Georgakoudi et al in two critical areas. First, we utilize an imaging approach which allows for faster screening in a clinical setting than point measurements. Arguably, point measurements require a visual target that can be interrogated for its spectral characteristics while an imaging approach can be used to visualize lesions not detectable by naked eye. While in this work the images of each lesion were binned, effectively creating a point-based measurement, the capability to acquire spectral images of sizable regions (>40 cm2) of tissue in clinically feasible times was demonstrated. The imaging approach can be easily implemented with the utilization of the most sensitive imaging techniques and classification methods, which was the focus of this work.

The second important difference relates to the wavelengths of the light signal used in this study for imaging which for all measurement were in the NIR spectral region. The benefits of using NIR light are widely recognized including larger penetration depth in tissues and reduced artifacts arising from absorption by blood. The nine imaging methods used in this work that are based in NIR light scattering take full advantage of these benefits. In addition, all three autofluorescence imaging methods were based in emission in the NIR and although two of these techniques utilized excitation at 408 and 532 nm, the analysis of the data suggested that 632.8 nm excitation (a wavelength where blood absorption is reduced) offered maximum sensitivity.

Combining wavelength regions such as those used by Georgakoudi et al. with the NIR wavelength regions used in this study would likely further improve the performance of the optical diagnostic. The addition of more light sources and optical modalities would likely improve the predictive capabilities of the system, but would likely increase the time required for interrogation. In addition, the trade-off between field-of-view and resolution translates into a significant increase in the amount of time required to screen larger areas at high resolution or point-based measurements. A method that may be ideal for surveillance screening of the esophagus may require adjustable field of view for margin delineation of suspect lesions. Such system may include multi-modal imaging and possibly point-based measurement capabilities for more accurate assessment of “regions of interest” that may be visualized by spectral imaging.

5. Conclusion

This study allowed the assessment of several near infrared optical imaging approaches for the pathological evaluation of esophageal biopsy specimens. By mathematically combining the data acquired via each imaging approach, it was possible to maximize the predictive capabilities of the separate optical methods. Furthermore, the combinations of all the imaging techniques were evaluated for maximal sensitivity and specificity. The results of this study showed that NIR optical techniques hold significant promise towards the future goal of minimally invasive, automated, real-time diagnosis of esophageal lesions.

Table 1. Pathologies of all the biopsy sets included in the study, with designated (a priori) risk category, and the predicted risk category as determined by logistic discriminant analysis of the combination of background-corrected mean intensities from the 1000 nm parallel- and cross-polarized, 633 nm fluorescence, 850 nm cross-polarized, and 700 nm degree-of-polarization images.

table-icon
View This Table

Acknowledgments

This work was performed in part at Lawrence Livermore National Laboratory under the auspices of the U.S. Department of Energy under Contract W-7405-Eng-48. This research is supported by the Center for Biophotonics, a National Science Foundation Science and Technology Center, managed by the University of California, Davis, under Cooperative Agreement No. PHY 0120999.

References and links

1.

American Cancer Society. Cancer Facts and Figures 2005, Atlanta: American Cancer Society (2005).

2.

M. Panjehpour, B. Overholt, T. Vo-Dinh, R. Haggitt, D. Edwards, and F. Buckley, “Endoscopic fluorescence detection of high-grade dysplasia in Barrett’s esophagus,” Gastroenterology 111, 93–101 (1996). [CrossRef] [PubMed]

3.

H. Tajiri, M. Kobayashi, K. Izuishi, and S. Yoshida, “Fluorescence endoscopy in the gastrointestinal tract,” Dig. Endosc. 12, S28–S31 (2000). [CrossRef]

4.

I. Georgakoudi, B. C. Jacobson, J. Van Dam, V. Backman, M. B. Wallace, M. G. Muller, Q. Zhang, K. Badizadegan, D. Sun, G. A. Thomas, L. T. Perelman, and M. S. Feld, “Fluorescence, reflectance, and light-scattering spectroscopy for evaluating dysplasia in patients with Barrett’s esophagus,” Gastroenterology 120, 1620–1629 (2001). [CrossRef] [PubMed]

5.

M. G. Shim, L. M. Song, N. E. Marcon, and B. C. Wilson, “In vivo near-infrared Raman spectroscopy: demonstration of feasibility during clinical gastrointestinal endoscopy,” Photochem. Photobiol. 72, 146–150 (2000). [PubMed]

6.

C. Kendall, N. Stone, N. Shepherd, K. Geboes, B. Warren, R. Bennett, and H. Barr, “Raman spectroscopy, a potential tool for the objective identification and classification of neoplasia in Barrett’s oesophagus,” J. Pathol. 200, 602–609 (2003). [CrossRef] [PubMed]

7.

M. B. Wallace, L. T. Perelman, V. Backman, J. M. Crawford, M. Fitzmaurice, M. Seiler, K. Badizadegan, S. J. Shields, I. Itzkan, R. R. Dasari, J. Van Dam, and M. S. Feld, “Endoscopic detection of dysplasia in patients with Barrett’s esophagus using light-scattering spectroscopy,” Gastroenterology 119, 677–682 (2000). [CrossRef] [PubMed]

8.

D. C. Pickard, I. J. Bigio, S. G. Bown, L. B. Lovat, P. M. Ripley, and M. Novelli, “Optical biopsy for the diagnosis of dysplasia in Barrett’s oesophagus,” in Optical Biopsy IV ,R. R. Alfono ed., Proc. SPIE 4613, 234–243, (2002). [CrossRef]

9.

J. S. Wang, J. S. Shi, Y. Z. Xu, X. Y Duan, L. Zhang, J. Wang, L. M. Yang, S. F. Weng, and J. G. Wu, “FT-IR spectroscopic analysis of normal and cancerous tissues of esophagus,” World J. Gastroenterol. 9, 1897–1899 (2003). [PubMed]

10.

Q. B. Li, X. J. Sun, Y. Z. Xu, L. M. Yang, Y. F. Zhang, S. F. Weng, J. S. Shi, and J. G. Wu, “Diagnosis of gastric inflammation and malignancy in Endoscopic biopsies based on Fourier transform infrared Spectroscopy,” Clin. Chem. 51, 346–350 (2005). [CrossRef] [PubMed]

11.

L. T. Perelman, V. Backman, M. Wallace, G. Zonios, R. Manoharan, A. Nusrat, S. Shields, M. Seiler, C. Lima, T. Hamano, I. Itzkan, J. Van Dam, J. M. Crawford, and M. S. Feld, “Observation of periodic fine structure in reflectance from biological tissue: A new technique for measuring nuclear size distribution,” Phys. Rev. Lett. 80, 627–630 (1998). [CrossRef]

12.

H. Fang, M. Ollero, E. Vitkin, L. M. Kimerer, P. B. Cipolloni, M. M. Zaman, S. D. Freedman, I. J. Bigio, I. Itzkan, E. B. Hanlon, and L. T. Perelman, “Noninvasive sizing of subcellular organelles with light scattering spectroscopy,” IEEE J. Sel. Top. Quantum Electron. 9, 267–276 (2003). [CrossRef]

13.

R. S. Gurjar, V. Backman, L. T. Perelman, I. Georgakoudi, K. Badizadegan, I. Itzkan, R. R. Dasari, and M. S. Feld, “Imaging human epithelial properties with polarized light-scattering spectroscopy,” Nat. Med. 7, 1245–1248 (2001). [CrossRef] [PubMed]

14.

S. G. Demos, R. Gandour-Edwards, R. Ramsamooj, and R. D. White, “Near-infrared autofluorescence imaging for detection of cancer,” J. Biomed. Opt. 9, 587–592 (2004). [CrossRef] [PubMed]

15.

S. G. Demos, R. Bold, R. D. White, and R. Ramsamooj, “Investigation of near infrared autofluorescence imaging for the detection of breast cancer,” IEEE J. Sel. Top. Quantum Electron. 11, 791–798 (2005). [CrossRef]

16.

M. T. Moesta, B. Ebert, T. Handke, D. Nolte, C. Nowak, W. E. Haensch, R. K. Pandey, T. J. Dougherty, H. Rinneberg, and P. M. Schlag, “Protoporphyrin IX occurs naturally in colorectal cancers and their metastases,” Cancer Res. 61, 991–999 (2001) [PubMed]

OCIS Codes
(170.3880) Medical optics and biotechnology : Medical and biological imaging
(170.6510) Medical optics and biotechnology : Spectroscopy, tissue diagnostics

ToC Category:
Medical Optics and Biotechnology

History
Original Manuscript: November 22, 2005
Revised Manuscript: March 6, 2006
Manuscript Accepted: March 8, 2006
Published: March 20, 2006

Virtual Issues
Vol. 1, Iss. 4 Virtual Journal for Biomedical Optics

Citation
Chad A. Lieber, Shiro Urayama, Nazir Rahim, Raymond Tu, Ramez Saroufeem, Boris Reubner, and Stavros G. Demos, "Multimodal near infrared spectral imaging as an exploratory tool for dysplastic esophageal lesion identification," Opt. Express 14, 2211-2219 (2006)
http://www.opticsinfobase.org/oe/abstract.cfm?URI=oe-14-6-2211


Sort:  Author  |  Year  |  Journal  |  Reset  

References

  1. American Cancer Society. Cancer Facts and Figures 2005, Atlanta: American Cancer Society (2005).
  2. M. Panjehpour, B. Overholt, T. Vo-Dinh, R. Haggitt, D. Edwards, and F. Buckley, "Endoscopic fluorescence detection of high-grade dysplasia in Barrett's esophagus," Gastroenterology 111, 93-101 (1996). [CrossRef] [PubMed]
  3. H. Tajiri, M. Kobayashi, K. Izuishi, and S. Yoshida, "Fluorescence endoscopy in the gastrointestinal tract," Dig. Endosc. 12, S28-S31 (2000). [CrossRef]
  4. I. Georgakoudi, B. C. Jacobson, J. Van Dam, V. Backman, M. B. Wallace, M. G. Muller, Q. Zhang, K. Badizadegan, D. Sun, G. A. Thomas, L. T. Perelman, and M. S. Feld, "Fluorescence, reflectance, and light-scattering spectroscopy for evaluating dysplasia in patients with Barrett's esophagus," Gastroenterology 120, 1620-1629 (2001). [CrossRef] [PubMed]
  5. M. G. Shim, L. M. Song, N. E. Marcon, and B. C. Wilson, "In vivo near-infrared Raman spectroscopy: demonstration of feasibility during clinical gastrointestinal endoscopy," Photochem. Photobiol. 72, 146-150 (2000). [PubMed]
  6. C. Kendall, N. Stone, N. Shepherd, K. Geboes, B. Warren, R. Bennett, and H. Barr, "Raman spectroscopy, a potential tool for the objective identification and classification of neoplasia in Barrett's oesophagus," J. Pathol. 200, 602-609 (2003). [CrossRef] [PubMed]
  7. M. B. Wallace, L. T. Perelman, V. Backman, J. M. Crawford, M. Fitzmaurice, M. Seiler, K. Badizadegan, S. J. Shields, I. Itzkan, R. R. Dasari, J. Van Dam, and M. S. Feld, "Endoscopic detection of dysplasia in patients with Barrett's esophagus using light-scattering spectroscopy," Gastroenterology 119, 677-682 (2000). [CrossRef] [PubMed]
  8. D. C. Pickard, I. J. Bigio, S. G. Bown, L. B. Lovat, P. M. Ripley, M. Novelli, "Optical biopsy for the diagnosis of dysplasia in Barrett's oesophagus," in Optical Biopsy IV, R. R. Alfono ed., Proc. SPIE 4613, 234-243, (2002). [CrossRef]
  9. J. S. Wang, J. S. Shi, Y. Z. Xu, X. Y Duan, L. Zhang, J. Wang, L. M. Yang, S. F. Weng, and J. G. Wu, "FT-IR spectroscopic analysis of normal and cancerous tissues of esophagus," World J. Gastroenterol. 9, 1897-1899 (2003). [PubMed]
  10. Q. B. Li, X. J. Sun, Y. Z. Xu, L. M. Yang, Y. F. Zhang, S. F. Weng, J. S. Shi, and J. G. Wu, "Diagnosis of gastric inflammation and malignancy in Endoscopic biopsies based on Fourier transform infrared Spectroscopy," Clin. Chem. 51, 346-350 (2005). [CrossRef] [PubMed]
  11. L. T. Perelman, V. Backman, M. Wallace, G. Zonios, R. Manoharan, A. Nusrat, S. Shields, M. Seiler, C. Lima, T. Hamano, I. Itzkan, J. Van Dam, J. M. Crawford, and M. S. Feld, "Observation of periodic fine structure in reflectance from biological tissue: A new technique for measuring nuclear size distribution," Phys. Rev. Lett. 80, 627-630 (1998). [CrossRef]
  12. H. Fang, M. Ollero, E. Vitkin, L. M. Kimerer, P. B. Cipolloni, M. M. Zaman, S. D. Freedman, I. J. Bigio, I. Itzkan, E. B. Hanlon, L. T. Perelman, "Noninvasive sizing of subcellular organelles with light scattering spectroscopy," IEEE J. Sel. Top. Quantum Electron. 9, 267-276 (2003). [CrossRef]
  13. R. S. Gurjar, V. Backman, L. T. Perelman, I. Georgakoudi, K. Badizadegan, I. Itzkan, R. R. Dasari, and M. S. Feld, "Imaging human epithelial properties with polarized light-scattering spectroscopy," Nat. Med. 7, 1245-1248 (2001). [CrossRef] [PubMed]
  14. S. G. Demos, R. Gandour-Edwards, R. Ramsamooj, and R. D. White, "Near-infrared autofluorescence imaging for detection of cancer," J. Biomed. Opt. 9, 587-592 (2004). [CrossRef] [PubMed]
  15. S. G. Demos, R. Bold, R. D. White, and R. Ramsamooj, "Investigation of near infrared autofluorescence imaging for the detection of breast cancer," IEEE J. Sel. Top. Quantum Electron. 11, 791-798 (2005). [CrossRef]
  16. M. T. Moesta, B. Ebert, T. Handke, D. Nolte, C. Nowak, W. E. Haensch, R. K. Pandey, T. J. Dougherty, H. Rinneberg, P. M. Schlag, "Protoporphyrin IX occurs naturally in colorectal cancers and their metastases," Cancer Res. 61, 991-999 (2001) [PubMed]

Cited By

Alert me when this paper is cited

OSA is able to provide readers links to articles that cite this paper by participating in CrossRef's Cited-By Linking service. CrossRef includes content from more than 3000 publishers and societies. In addition to listing OSA journal articles that cite this paper, citing articles from other participating publishers will also be listed.


« Previous Article  |  Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited