Search

Background and ObjectivesEndovenous laser treatment (ELT) has been recently proposed as an alternative in the treatment of reflux of the great saphenous vein (GSV) and small saphenous vein (SSV). Successful ELT depends on the selection of optimal parameters required to achieve an optimal vein damage while avoiding side effects. Mathematical modeling of ELT could provide a better understanding of the ELT process and could determine the optimal dosage as a function of vein diameter.Study design–materials and methodsThe model is based on calculations describing the light distribution using the diffusion approximation of the transport theory, the temperature rise using the bioheat equation and the laser-induced injury using the Arrhenius damage model. The geometry to simulate ELT was based on a 2D model consisting of a cylindrically symmetric blood vessel including a vessel wall and surrounded by an infinite homogenous tissue. The mathematical model was implemented using the Macsyma-Pdease2D software (Macsyma Inc., Arlington, MA, USA). Damage to the vein wall for CW and single shot energy was calculated for one vein diameter (3 mm). In pulsed mode, the pullback distance (3, 5 and 7 mm) was considered. For CW mode simulation, the pullback speed (1, 2, 3 mm/s) was the variable. The total dose was expressed as joules per centimeter in order to perform comparison to results already reported in clinical studies.ResultsIn pulsed mode, for a 3 mm vein diameter, irrespective of the pullback distance (2, 5 or 7 mm), a minimum fluence of 15 J/cm is required to obtain a permanent damage of the intima. In continuous mode, for a 3 mm vein diameter, 65 J/cm are required to obtain a permanent damage of the vessel wall. Finally, the use of different wavelengths (810 or 980 nm) played only a minor influence on these results.Discussion and conclusionThe parameters determined by mathematical modeling are in keeping with those used in clinical practice. They confirm that thermal damage of the inner vein wall is required to obtain a permanent occlusion. However, in order to obtain a high rate of success, without adverse events, the knowledge of the vein diameter after tumescent anesthesia is required. When performing ELT, the pulsed mode should be preferred for 2 reasons: 1) it requires less energy since the vein is only damaged sequentially and not continuously, 2) it reduces the risk of too high an energy inside the vein when pullback speed is not controlled. So, the pulsed mode requires a very precise positioning of the fiber after each pullback and the time of treatment is longer but it has been prove that this technique is reproducible, safe and efficient [29].

The intensity noise properties of Brillouin fiber ring lasers are investigated both theoretically and experimentally. The fluctuating parameters that have the dominant influence on laser intensity noise are the pump rate and the cavity reinjection rate. The transfer functions that relate the laser intensity noise to the fluctuations of these parameters are determined in a theoretical study that is performed within the framework of the usual three-wave model of stimulated Brillouin scattering. The theoretical predictions are confirmed by experiments performed in a Brillouin fiber ring laser operating in a low-finesse cavity. Finally, the ability of all-fiber Brillouin lasers to reduce intensity noise of pump sources is discussed.

We study, both theoretically and experimentally, the existence and behavior of surface localized modes in a one-dimensional (1D) comblike photonic band gap structure. The structure is composed of a waveguide, along which dangling side branches are grafted periodically. We demonstrate a theoretical rule about the existence of surface states in such 1D periodic structures; namely, when one considers together two semi-infinite structures obtained by cutting an infinite one along a plane perpendicular to the axis of the structure, one always has as many localized surface states as minigaps. We propose an original method to observe the surface modes, namely, by measuring the transmission through a guide along which a finite comb is attached vertically. The surface modes of the comb are obtained as maxima or minima of the transmission spectrum. The experiments are performed by using coaxial cables in the frequency range of a few hundred MHz.

Abstract:: We analyse the hysteresis enlargements of an optical bistable system involving three dynamical variables. We investigate, both experimentally and numerically, the local dynamics of the up- and down-switching process versus the sweeping frequency of the control parameter. In particular, we delineate the domain of validity of the scaling law predicted for one-dimensional systems. At high sweeping frequency, we show the appearance of another asymptotic scaling low in . Thereafter, we analyse the global evolution of the hysteresis loop induced by these processes. At low frequency, a scaling law is retrieved, whereas at high frequency, the dynamical behaviour is shown to strongly depend on the particular shape of the bistability curve.

Nous discutons quelques mécanismes d'instabilité récemment observés dans un oscillateur paramétrique optique (OPO) : d'une part des instabilités opto-thermiques où le système oscille autour des courbes de résonance d'un ou plusieurs modes, d'autre part des oscillations rapides résultant de l'interaction de plusieurs modes transverses. La première observation expérimentale de chaos déterministe dans un OPO est également présentée.

Les caractéristiques du bruit d'intensité dans un laser Brillouin à fibre sont étudiées expérimentalement et théoriquement. Les fluctuations d'intensité de ce laser proviennent de deux sources de bruit : le bruit du laser de pompe et les fluctuations du coefficient de réinjection de la cavité. Le transfert du bruit d'une source sur l'intensité du laser Brillouin est défini par une fonction de gain qui a été déterminée expérimentalement et théoriquement.

De récentes expériences ont mis en évidence l'apparition d'un trou dans la courbe de gain de l'amplificateur Brillouin [1] et dans celle du générateur Brillouin [2]. Ce comportement est généralement interprété comme étant un effet lié à un élargissement inhomogène du gain. Pourtant, nous montrons que ce trou n'est aucunement lié à l'élargissement inhomogène mais qu'il est descriptible dans le cadre d'un gain homogène.