Modeling laser damage to the retina
This dissertation presents recent progress in several areas related to modeling laser damage to the retina. In Chapter 3, we consider the consequences of using the Arrhenius damage model to predict the damage thresholds of multiple pulse, or repetitive pulse, exposures. We have identified a few fundamental trends associated with the multiple pulse damage predictions made by the Arrhenius model. These trends differ from what would be expected by non-thermal mechanisms, and could prove useful in differentiating thermal and non-thermal damage.
Chapter 4 presents a new rate equation damage model hypothesized to describe photochemical damage. The model adds a temperature dependent term to the simple rate equation implied by the principle of reciprocity that is characteristic of photochemical damage thresholds. A recent damage threshold study, conducted in-vitro, has revealed a very sharp transition between thermal and photochemical damage threshold trends. For the wavelength used in the experiment (413 nm), thermal damage thresholds were observed at exposure levels that were twice the expected photochemical damage threshold, based on the traditional understanding of photochemical damage. Our model accounts for this observed trend by introducing a temperature dependent quenching, or repair, rate to the photochemical damage rate. For long exposures that give a very small temperature rise, the model reduces to the principle of reciprocity. Near the transition region between thermal and photochemical damage, the model allows the damage threshold to be set by thermal mechanisms, even at exposure above the reciprocity exposure.
In Chapter 5, we describe a retina damage model that includes thermal lensing in the eye by coupling beam propagation and heat transfer models together. Thermal lensing has recently been suggested as a contributing factor to the large increase in measured retinal damage thresholds in the near infrared. The transmission of the vitreous decreases significantly for wavelengths in the near infrared due to an increase in the absorption coefficient for these long wavelengths. This means that less energy actually reaches the retina, but it also means that more energy is absorbed by the vitreous which can lead to significant temperature rises. The refractive index of water is known to depend on temperature, and the vitreous has very similar optical properties to water, so temperature gradients in the vitreous lead to refractive index gradients that act as a lens. Since the refractive index of water decreases with an increase in temperature, the overall effect is to establish a negative lens that defocuses a beam inside the eye during a laser exposure. This effect is a potential protection mechanism for the retina, as it would limit the time for which a laser can be sharply focused on the retina. Our model agrees well with thermal lensing measurements that have been conducted in water and we have used it to predict the retinal damage threshold as a function of exposure duration for 1318 nm exposures at various beam diameters. The model predicts that the damage threshold remains constant after some exposure time, which depends on the beam diameter. This is due mainly to the fact the retinal temperature rise is limited by the thermal lens and reaches a peak value in a relatively short time (on the order of 10 ms), which limits the amount of time that a laser exposure can cause damage.
Finally, in Chapter 6 we describe the first steps we have taken in building a comprehensive short pulse retina damage model. Currently, no model capable of predicting retinal damage outcome based on the exposure parameters at the cornea exists. Models of possible damage mechanisms do exist (the damage mechanism for pulses less than about 1 mus are non-thermal), but these models assume that the exposure parameters are known at the absorption site (the retina). We have constructed a configurable, linear short pulse propagation model, that is capable of predicting the exposure to the retina from a pulse incident on the cornea. The model includes all of the linear effects of scalar wave theory such as group velocity dispersion, higher-order dispersion, aberration and even a wavelength dependent absorption coefficient. The model will allow the retina exposure to be calculated from a wide range of cornea exposure configurations, which can then be used as input into various short pulse damage models.