No abstract

Raman spectroscopy (RS) has potential for disease classification within the gastrointestinal tract (GI). A near-infrared (NIR) fiber-optic RS system has been developed previously. This study reports the first in vivo Raman spectra of human gastrointestinal tissues measured during routine clinical endoscopy. This was achieved by using this system with a fiber-optic probe that was passed through the endoscope instrument channel and placed in contact with the tissue surface. Spectra could be obtained with good signal-to-noise ratio in 5 s. The effects on the spectra of varying the pressure of the probe tip on the tissue and of the probe-tissue angle were determined and shown to be insignificant. The limited set of spectra from normal and diseased tissues revealed only subtle differences. Therefore, powerful spectral-sorting algorithms, successfully implemented in prior ex vivo studies, are required to realize the full diagnostic potential of RS for tissue classification in the GI.

Recent reports indicate that useful diagnostic information can be obtained from ex vivo tissue by using near-infrared Raman spectroscopy. A fiber-optic-based Raman system has been constructed that can obtain spectra in vivo from tissue, typically in less than 30 s. The spectral details are limited by the fluorescence and Raman signal generated in the silica delivery and collection fibers. In this study, fiber-optic probes that are designed to suppress these confounding signals have been characterized and compared with a probe constructed of unfiltered silica fiber. These suppressing probes used a novel design with “in-the-tip” filters, which also had optimized light collection efficiency. The spatial point response functions of the fibers were measured in air, water, and a tissue-simulating, optically turbid medium. Spectra from rabbit tissues were also collected with these probes to demonstrate their performance.

Recent studies in the literature have investigated the feasibility of tissue diagnostics based on Raman spectroscopy. The majority of these compare the ex vivo spectra of normal and diseased tissue. Due to the time lapse between tissue excision and spectroscopic examination, samples must be frozen or otherwise preserved to maintain their native biochemical states. In order to establish optimum procedures for ex vivo Raman spectroscopy of tissue, the effects of tissue drying, formalin fixing, snap freezing, tissue freezing in optimal cutting temperature (OCT) medium and extended post-thaw durations were studied to determine if any of these handling procedures introduced spectral artifacts. Experiments on representative tissues indicated that tissue heating due to the excitation light did not change the spectra significantly. With minor exceptions, OCT and formalin did not contaminate tissue spectra, so that samples stored for histological examination could also be studied with Raman spectroscopy. Tissue dehydration caused disruption of the protein vibrational modes, which caused spectral artifacts. It is concluded that ex vivo tissue samples should be frozen in OCT. Prior to spectral analysis, the tissue should then be acclimatized at room temperature in phosphate-buffered saline (PBS) and immersed in PBS during spectroscopic examination.

Raman spectroscopy (RS) has potential for disease classification within the gastrointestinal tract (GI). A near‐infrared (NIR) fiber‐optic RS system has been developed previously. This study reports the first in vivo Raman spectra of human gastrointestinal tissues measured during routine clinical endoscopy. This was achieved by using this system with a fiber‐optic probe that was passed through the endoscope instrument channel and placed in contact with the tissue surface. Spectra could be obtained with good signal‐to‐noise ratio in 5 s. The effects on the spectra of varying the pressure of the probe tip on the tissue and of the probe‐tissue angle were determined and shown to be insignificant. The limited set of spectra from normal and diseased tissues revealed only subtle differences. Therefore, powerful spectral‐sorting algorithms, successfully implemented in prior ex vivo studies, are required to realize the full diagnostic potential of RS for tissue classification in the GI.

No abstract

No abstract

The purpose of this study is to assess if Raman spectroscopy can classify dysplastic (DYS) and early neoplastic lesions within Barrett's esophagus (BE). In BE, the normal squamous epithelium (SQ) lining the esophagus is replaced by columnar epithelium (intestinal metaplasia). These patients have a 30-125 fold excess risk of developing adenocarcinoma. Raman spectroscopy may provide diagnostic information so that tissue transformation may be detected at an early stage (dysplasia) and improve the patient's prognosis. Ex vivo measurements were carried out initially on biopsy samples obtained from BE patients undergoing routine endoscopic and biopsy surveillance. Differences were noted in the spectral regions 1200-1350 cm and 1550-1640 cm1 when comparing different histopathologic grades. Principal component analysis of the spectra led to good separation between SE and BE but not between BE and DYS. Improved results were obtained using a probabilistic artificial neural network, with a resultant sensitivity and specificity of 77% and 93% in differentiating SQIBE from dysplasia, respectively. Recently, in vivo endoscopic measurements have been performed. These preliminary results indicate that RS in combination with endoscopy may be a useful technique to screen BE patients for dysplastic/early neoplastic lesions.