Partners and International Organizations
(English)
|
Dr. H. Sterenborg (AMC, Amsterdam, NL), Dr. R. Baumgartner (LFUK, München, D), Prof. W. Star (DdH, Rotterdam, NL), Dr. S. Mordon (URTB, Lille, F), Dr. K. Irion (Storz, Tuttlingen, D), Prof. A. Ayache (INP, Toulouse, F), Dr. M. Canis (CLER, Clermont Ferrand, F), Dr. N. van der Vange (AVL, Amsterdam, NL)
|
Abstract
(English)
|
Fluorescence imaging is a useful tool to detect early stage cancers as well as pre-cancerous lesions in hollow organs. The principle of this method is to detect an optical contrast between the deseased tissue and the healthy surroucding area. Up to now, essentially all the works performed in this field were based on steady-state fluorescence images where this contrast is based on the intensity and/or the spectroscopy of the fluorescence. Another approach is to measure the fluorescence lifetime (FLT) contrast. This new approach provides valuable information to understand the mechanisms responsible for the fluorescence intensity/spectroscopy changes associated with the presence of the lesions. Importantly, it is potentially more sensitive and specific to detect early cancers than other techniques. We drove our choice to the so called 'frequency-domain FLT imaging' (FD-FLIM) approach. Principles of FD-FLIM consist in modulating the excitation light at a radio frequency and measuring the phase-shift or the demodulation of the fluorescence induced by this modulated light. The phase-shift and demodulation occur due to the FLT of the fluorochrome that acts as a first-order low-pass filter. To adapt this principle to imaging techniques for endoscopic examination and to assess the performances of this method to detect and characterize early cancerous lesions of the cervix is the main purpose of this work.
The planning of this research has been divided into three main parts. The first one consisted in validating the principle to the field of endoscopic imaging. It is based on computer simulation (stochastic computing) using input parameters (tissue optical parameters, features and principles of the instrumentation) obtained from separate experiments. The second one has been the realization of an apparatus able to produce these lifetime images of the cervix, to test it with ex-vivo tissue samples, to optimize the excitation mode (pulsed versus sine-wave excitation) as well as the data acquisition and processing (improvement of the signal to noise ratio by a factor 1.7), and to characterize the photonic features of this instrumentation. The third one consisted to perform the in-vivo clinical studies.
Nineteen patients have been involved in this study which started in June 2000 at the Geneva University Hospital. Three of them presented low grade squamous intraepithelial lesions (SIL) whereas only one presented high grade SIL. The mean FLT is 1.5 ns on the lesions before acetic acid application and 1.7 ns after acetic acid application. The precision of the instrumentation is about ±25%. The application of acetic acid also broadens the lifetime distribution. No temporal contrast has been observed between dysplastic and healthy tissues with the spectral design used for this study (Ex.: 532 nm; Em.: 550-700 nm; Modul.: 40 MHz).
One part of the results and achievements mentioned above are going to be at the origin of the publication of a second EPFL thesis manuscript (author: Th. Stepinac) in several months.__
|
