| Abstract |
The first scope of this work was to investigate the effects of liquid nitrogen and formalin-based conservation in the laser-induced fluorescence spectra taken from peripheral vascular tissue. A He-Cd laser emitting at 442nm and an Ar+ laser emitting at 457.9nm, 476.5nm, 488nm, 496.5nm and 501.7nm were used as excitation sources. Separate sets of measurements were made for all samples at each of these wavelengths. All samples were obtained from by-pass operations and amputations performed at the Vascular Surgery Clinic of the University of Crete. The samples were irradiated one hour after the excision. The fluorescence spectra from fresh tissues were compared to those taken after the tissues were stored in liquid nitrogen or formalin for 24 and 48 hours. The comparison of the fluorescence signals was made with the implementation of twelve (12) simple algebraic algorithms which were based on the intensity difference of the recorded spectra. All the in vitro laser induced fluorescence spectroscopy studies were performed after the conservation of the specimens for a certain time in formalin or in liquid nitrogen. So far the effects of the liquid nitrogen or formalin-based conservation in the fluorescence spectra of the tissues have not been extensively studied. The lack of this kind of study in the international literature made its realization a necessity. The aims of this study were: the investigation of the changes in the fluorescence spectra due to conservation, the study of the time required for the generation of such changes and the selection of a way of conservation which induces the less changes in the distribution of the recorded fluorescence signals. The algorithms seemed to give better results in the discrimination of the tissues, for formalin than for liquid nitrogen conservation. Some of the algorithms succeeded to detect changes in the fluorescence signal of tissue samples obtained directly after excision from those stored in liquid nitrogen or formalin for 24 hours, but usually failed to detect any variations between 24 and 48 hours. The results suggest that liquid nitrogen conservation does not seriously alter the spectral features of excised tissue whereas formalin conservation seem to affect the distribution of the laser induced fluorescence spectra within the first 24 hours. The second scope of this work was to investigate the feasibility of application of laser induced fluorescence spectroscopy (in vitro) in order to discriminate between normal and pathologic peripheral vascular tissue. A He-Cd laser emitting at 442nm and an Ar+ laser emitting at 457.9nm, 476.5nm, 488nm, 496.5nm were again used as excitation sources. All samples were subjected to either single or dual wavelength excitation from the He-Cd laser and one of the Ar+ laser emission lines. All the samples (abdominal aortas, as well as flank, femoral, tibial, fibular and ham artery tissues) were obtained from by-pass operations and amputations performed at the aforementioned Clinic. The discrimination and the classification of different types of tissue (normal artery, fibrous plaque, calcified plaque) was done with: i) the implementation of twelve (12) simple algebraic algorithms, which were related to the recorded spectra and ii) study of the full width at half maximum (FWHM) of the spectra. In addition, a number of the obtained fluorescence spectra was processed with the use of Artificial Neural Networks (ANN) in order to analyze them in a more detailed way. Furthermore, the way of excitation (single or dual wavelength) which gives the best results in terms of discrimination of tissues was investigated. Finally, a number of in vivo fluorescence measurements was taken during by-pass operations using as an excitation source a He-Cd laser. The in vivo measurements were performed at The Royal London Hospital, Whitechapel, London. The primary aim of this study, was to investigate the possibility for the clinical use of this relatively new, minimally invasive technique (laser-induced fluorescence spectroscopy), as a diagnostic tool for the early detection of diseased peripheral vascular tissues. The collection and the analysis of the data was made in a relatively small time interval (a few seconds), in order to make possible the application of this technique to in vivo measurements too. The intention is to use this technique in conjunction with the existing diagnostic methods, in order to obtain more accurate and reliable information about diseased peripheral vascular tissues at an early stage. In the case of in vitro measurements, the use of algorithms made possible the discrimination between normal and fibrous or calcified tissue. The application of the algorithms seemed to be more efficient regarding the discrimination of tissues in the case of dual wavelength excitation. This was more profound in the discrimination between normal and fibrous tissue. The best results were obtained when the combination of wavelengths 442nm + 488nm and 442nm + 496.5nm were used for the excitation of the samples. The analysis of the fluorescence spectra with Artificial Neural Networks (ANN) allowed not only the discrimination between normal and fibrous and normal and calcified tissue, but the discrimination between fibrous and calcified plaque as well. For the in vivo measurements the samples were subjected only to single wavelength excitation with the use of the He-Cd laser (442nm). Initial results were promising with respect to the discrimination between normal and abnormal vessels. Nevertheless, further studies are needed to support the preliminary observations. The third scope of this work was to investigate fluorescence spectra taken from lamb and human hearts (in vitro). An Ar+ laser emitting at 457.9nm was used for excitation. The lamb hearts were irradiated within the first two hours after the excision. The human hearts were obtained from the morgue and were irradiated one hour after the excision. Spectra from different cardiac compartments (the left and right atria and ventricles, the myocardium, the epicardium, the aorta and the fat deposits) were recorded. It was investigated whether each chamber exhibited constant spectral response. Efforts were also made to discriminate the different cardiac compartments, using laser-induced fluorescence spectroscopy, by comparing the spectral intensity and by using simple algorithms based on the spectral intensity variation. The alterations in the fluorescence spectra due to the conservation of the samples in formalin for 48 hours were also investigated. Finally, a number of in vivo measurements was taken during open heart operations at the Saint Bartholomew Hospital, London. Due to technical reasons, these measurements were performed using the He-Cd laser. In the present work healthy hearts were examined. In this way the emission spectrum of the myocardial tissue was investigated and an effort was made to detect any pathologic state of the myocardium. The aim is to use this diagnostic technique along with the ordinary biopsy, for the extraction of better results. It was observed that both in human and in lamb heart tissue all the measurements that were taken from the same chamber were remarkably constant. Since the samples were characterized as healthy, it is expected that there will be alterations in the fluorescence signals recorded from diseased parts. In all the hearts that were investigated (both human and lamb), the fluorescence spectra obtained from the aortas and the atria (in particular the left atrium) were more intense than the ones obtained from other parts (e.g. ventricles). The fluorescence spectra recorded from the myocardium of the human heart (in vitro) exhibited significant spectral differences from the ones recorded from other compartments of the heart. After the conservation of the human and lamb hearts in formalin for 48 hours, changes were observed in their spectral response, similar to those observed in the case of the peripheral vascular tissues. The fluorescence signals that were obtained during the in vivo measurements in human hearts presented a lot of similarities (in the morphology and in the change of the intensity of the signal from different cardiac compartments) with the spectra that were recorded in vitro. The methodology and the results of this work are expected to have clinical applications in the future of the diagnosis of cardiovascular diseases.
|
Collections
