Robert M. Cothren

Gastrointestinal tissue diagnosis by laser-induced fluorescence spectroscopy at endoscopy

An endoscope-compatible, optical fiber system has been developed which can be used to obtain laser-induced fluorescence spectra of mucosal abnormalities during endoscopy in real time. The results of our previous in vitro studies have suggested that laser-induced fluorescence tissue spectra are sufficiently unique that they can be used to accurately diagnose mucosal abnormalities in some systems. To test this hypothesis in vivo, laser-induced fluorescence spectra were obtained during colonoscopy from 31 colonic adenomas, 4 hyperplastic polyps, and 32 examples of normal mucosa in 20 patients. The resulting spectra could be used to correctly differentiate adenomas from normal colonic mucosa and hyperplastic polyps in 97% of the specimens studied with the resulting sensitivity, specificity, and positive predictive value of 100%, 97%, and 94%, respectively. These results, although preliminary in nature, suggest that laser-induced fluorescence spectra can be used in the recognition and differential diagnosis of mucosal abnormalities at endoscopy.

Detection of dysplasia at colonoscopy using laser-induced fluorescence: a blinded study

BACKGROUND Laser-induced fluorescence spectroscopy has the potential to detect colonic dysplasia in vivo. However, previous studies have limited their analyses to multivariate regression techniques and unblinded retrospective evaluation. The purpose of this study was to develop a probability-based algorithm to detect colonic dysplasia using laser-induced fluorescence spectroscopy and to evaluate it in a blinded manner. METHODS Fluorescence spectra were collected from normal mucosa and colonic polyps during colonoscopy using 370 nm excitation. Tissue was classified as normal, hyperplastic, or adenomatous by histologic examination. Preliminary data was used to devise an algorithm to differentiate tissue type based on probability distributions of the fluorescence intensity at 460 nm and the ratio of the intensity at 680 nm to that at 600 nm. The algorithm was then tested in a blinded fashion. RESULTS The algorithm correctly determined the tissue type in 88% of cases, equal to the agreement of independent pathologists. Sensitivity, specificity, and positive predictive value for the detection of dysplasia was 90%, 95%, and 90%, respectively. CONCLUSIONS Dysplasia was detected in vivo using fluorescence spectroscopy and a probability-based algorithm. This method may form the basis for a new surveillance technique for patients with increased risk for dysplastic transformation.

Spectral diagnosis of atherosclerosis using an optical fiber laser catheter.

This communication demonstrates that fluorescence spectra of human aorta with good S/N ratios can be collected using an optical fiber laser catheter. The performance of this catheter is compared to a non-fiber optic collection system with an equivalent delivery/collection geometry. For a given sample, fluorescence lineshapes obtained using the two systems are identical; differences in peak fluorescence intensity are related to the different collection efficiencies of the two systems. It is shown that the fluorescence lineshape of arterial tissue depends on the delivery/collection geometry of the detection system, and that this is due to the interaction of absorption and fluorescence within the artery wall. This effect is investigated systematically using a specially designed collection system. Results are analyzed qualitatively using a simple, one-dimensional model of tissue fluorescence. With this analysis, we present design requirements for a collection system in which such geometric effects are eliminated, and show that our optical fiber laser catheter satisfies these requirements.

Three-dimensional segmentation of luminal and adventitial borders in serial intravascular ultrasound images

Intravascular ultrasound (IVUS) provides exact anatomy of arteries, allowing accurate quantitative analysis. Automated segmentation of IVUS images is a prerequisite for routine quantitative analyses. We present a new three-dimensional (3D) segmentation technique, called active surface segmentation, which detects luminal and adventitial borders in IVUS pullback examinations of coronary arteries. The technique was validated against expert tracings by computing correlation coefficients (range 0.83-0.97) and Williams index values (range 0.37-0.66). The technique was statistically accurate, robust to image artifacts, and capable of segmenting a large number of images rapidly. Active surface segmentation enabled geometrically accurate 3D reconstruction and visualization of coronary arteries and volumetric measurements.

A model for extraction of diagnostic information from laser induced fluorescence spectra of human artery wall

Abstract We have developed a method of analyzing fluorescence signals from optically thick tissue, which can be used to extract important clinical information about tissue type and chemical composition. Our model describes the composite tissue spectrum in terms of contributions from fluorophores and absorbers which modulate the intrinsic fluorescence spectrum. Scattering is included empirically in the model by describing absorbing species with ‘attenuation spectra’, which include contributions from both absorption and scattering. We present an analysis of laser induced fluorescence data obtained from normal and atherosclerotic human artery wall in vitro . We show that the model can be used to separate effects due to fluorescence and absorption, and to deconvolute contributions from individual chromophores, extracting information about their relative concentrations. This information can be utilized to diagnose tissue type more accurately than empirical algorithms, and may potentially prove to be useful in determining the chemical composition of atherosclerotic lesions in vivo .

Automated morphometry of coronary arteries with digital image analysis of intravascular ultrasound.

We designed and tested digital image processing strategies to perform fully automated segmentation of luminal and medial-adventitial boundaries in intravascular ultrasound images of human coronary arteries. Automated segmentation is an essential tool for advanced techniques of clinical visualization and quantitative measurement. Vascular compliance measurements and three-dimensional reconstructions are demonstrated as examples of such applications. Digital image processing was performed in three phases: (1) preprocessing, including a polar transform, local contrast enhancement, and speckle noise filtering; (2) segmentation, involving radial scanning, region growing, or cost-function minimization techniques; and (3) postprocessing, involving dropout filtering and outline smoothing. Cross-sectional areas were compared with manual tracings from experienced operators and showed good agreement. The algorithm bias ranged from -0.34 to 1.18 mm2; interclass and intraclass correlation coefficients ranged from 0.83 to 0.94. The designed techniques currently allow fully automated segmentation without operator interaction of the luminal and, if present, medial-adventitial boundary.

Three-dimensional reconstruction of the coronary artery wall by image fusion of intravascular ultrasound and bi-plane angiography

Background: Intravascular ultrasound (IVUS) is becoming increasingly accepted for assessing coronary anatomy. However, its utility in visualizing and quantifying coronary morphology has been limited by its 2D tomographic nature. This study presents a 3D reconstruction technique that accurately preserves 3D geometric information. Methods and Results: Images obtained from manual IVUS pullbacks and continuous bi-plane angiography were fused, using angiography to reconstruct the transducer trajectory and aid in solving for the correct rotational orientation. A novel 3D active surface method automatically identified the luminal and medial–adventitial borders which, when superimposed on the transducer trajectory, could be surface-rendered for visualization and morphometry. Segmentation agreed well with manual assessment, and 3D luminal shape matched that of angiography when projected to 2D. Conclusions: We conclude that this method provides an accurate reconstruction of the vessels anatomy, which accounts for the true curvature of the vessel.

Characterization of ultraviolet laser-induced autofluorescence of ceroid deposits and other structures in atherosclerotic plaques as a potential diagnostic for laser angiosurgery☆☆☆

Unstained frozen sections of normal and atherosclerotic human aorta and coronary artery were examined using histochemical and fluorescence microscopic techniques to identify the structures responsible for autofluorescence under 351 to 364 nm laser excitation. These structures included elastin and collagen in normal and atherosclerotic specimens, calcium deposits in calcified plaques, and granular or ring-shaped deposits histochemically identified as ceroid found in both calcified and non-calcified plaques. Qualitatively, both the color and intensity of ceroid autofluorescence differed greatly from that of elastin or collagen. The emission spectra of elastin, collagen, and ceroid were examined by microscopic spectrofluorimetry, and were found to differ significantly as well. When compared with spectra of elastin and collagen, spectra of ceroid were broader, shifted to the red, and were somewhat resistant to bleaching. We conclude that detection of laser-induced ceroid autofluorescence may aid in identifying plaques for laser ablation.

Frequency content of normal and diabetic plantar pressure profiles: Implications for the selection of transducer sizes

How small do pressure transducers need to be in order to faithfully measure the plantar pressure profiles (PPPs) under normal and diabetic feet? In this study, pressures were collected from five diabetic and six non-diabetic subjects using a commercial measurement system with 25 mm2 transducers. Discrete Fourier Transform techniques were then used to determine (i) the spatial frequency content of diabetic and non-diabetic PPPs, and (ii) the effects of quadrupling the transducer area (from 5 mm x 5 mm to 10 mm x 10 mm). When the data were filtered to represent the effects of using 10 mm x 10 mm transducers, it was found that the ensuing reductions in peak pressure in the toe region (50 kPa) were significantly greater than in all other regions of the foot (p < 0.05). There was a significant correlation between pressure underestimations and measured peak pressures in the metatarsal regions. Based on data collected with 25 mm2 transducers it was concluded that transducer sizes greater than 6.36 mm x 6.18 mm (medio-lateral and antero-posterior directions) would result in sub-optimal sampling of PPPs.

Preliminary evaluation of the prototype stereoscopic endoscope: precise three-dimensional measurement system

A prototype stereoscopic endoscope (incorporating two charged-coupled devices), developed for the accurate three-dimensional measurement of gastrointestinal tract lesions, was initially evaluated with two-dimensional target grids and in vitro measurement of 15 objects of known size (marbles, cubes, and rectangular prisms) placed in a plastic model of the sigmoid colon. Images of the objects were captured and stored in a computer. Stereoscopic measurements were compared with results from the standard (open biopsy forceps) method by a blinded endoscopist. The volume measured with the stereoscope did not differ significantly from the true volume, whereas the volume obtained with the open biopsy forceps method differed significantly from the actual volume, consistently underestimating the actual size. The aberration ratios (percentage deviation between the measured and true volume, expressed as mean +/- SD) obtained with the stereoscopic endoscope were superior (9.2% +/- 9.5%) to those obtained with the open biopsy forceps method (-34.0% +/- 26.8%). These preliminary in vitro results with the stereoscope show considerable promise for the simple and precise three-dimensional measurement of gastrointestinal lesions and warrant human clinical trials.