China has a high prevalence of eye diseases, with retinal conditions, such as diabetic retinopathy and macular degeneration, posing significant threats to vision. In current clinical practice, laser therapy is commonly used to treat these retinal diseases. However, improper selection of laser parameters during treatment can lead to adverse effects, such as vision damage, in over half of the cases. Therefore, it is crucial to develop a heat transfer model for laser therapy of retinal diseases to assist physicians in adjusting laser parameters and formulating appropriate treatment plans for different patients. Previous models of eye have led to various simplifications and cannot accurately simulate the multiscale intraocular heat transfer processes from the macroscopic whole eye to the microscopic retina during laser treatment. To address this issue, in this study, a coupled bioheat transfer model of the human eye is developed, providing a basis for selecting laser parameters in retinal surgeries.

In this study, a coupled bioheat transfer model of human eye is developed. For the heat transfer process at the whole eye scale, Penne bio-heat transfer equation is employed. For the microscopic retinal scale, different parts of the fundus are modeled as porous media comprising biological tissues mixed with chromophores, and a two-temperature equation is established to represent the non-equilibrium heat transfer between the chromophore tissue and surrounding tissue matrix. By combining Penne bio-heat transfer equation with a two-temperature model, this model enables the simulation of laser retinal surgery across different time scales. In this model, calculations are performed using ANSYS software. First, a 3D geometric model of the entire eye is created using SolidWorks, followed by the division of the fully coupled eye model into polyhedral meshes using Fluent Meshing. Fine meshes are applied to narrow tissue structures, such as cornea, retina, choroid, and sclera, while coarser meshes are used for thicker tissue structures such as vitreous body and lens. The finite volume method in the ANSYS Fluent 2022 solver is used to solve the discretized equations. A double-precision coupled solution approach with a second-order implicit scheme is adopted, and the energy equation is solved using a second-order upwind scheme. After grid independence verification, the final model employs 7.02 million polyhedral cells.

The results show that thermotherapy via transpupillary thermotherapy can easily lead to retinal damage due to its longer treatment time. This in turn increases the risk of local recurrence and scleral extension, often resulting in suboptimal surgical outcomes. Panretinal photocoagulation can effectively heat the three light-absorbing layers of the fundus under quasi-thermal equilibrium conditions, causing thermal coagulation of damaged retinal tissue. However, this approach can damage retinal photoreceptors, potentially leading to scarring and anatomical disruption of the retina. Subthreshold diode micropulse, which uses lower laser energy, only thermally stimulates the release of cytokines from the retinal pigment epithelium (RPE) layer without causing significant damage to the retinal photoreceptors. Therefore, it is an effective treatment for retinal diseases such as macular degeneration.

This study addresses the issue that existing models are still unable to accurately simulate the thermal process of multi-scale lesion targets in retinal laser surgery. A fully coupled model of the entire eye has been developed, which can simulate the thermal processes of retinal laser surgery across the full time scale, from microseconds to seconds. The model couples Pennes bioheat transfer equation with a two-temperature non-equilibrium heat transfer model in porous media to calculate the thermal effects of typical retinal laser surgeries, including transpupillary thermotherapy (TTT), panretinal photocoagulation (PRP), and subthreshold micropulse diode laser (SDM) therapy. This model aids in optimizing laser surgery parameters for retina. The simulation results indicate that TTT, due to its longer treatment time, sometimes leads to retinal damage, increasing the risk of local recurrence and scleral extension, often resulting in poor surgical outcomes. Furthermore, PRP effectively heats the three light-absorbing layers of the fundus under quasi-thermal equilibrium conditions, but it may cause damage to adjacent retinal matrix tissues (photoreceptors), leading to night vision loss, scarring, and anatomical disruption of the retina. In SDM therapy, the use of lower energy only stimulates the release of cytokines from the RPE cells during recovery, without causing significant damage to the retinal nerves, making it an effective treatment for retinal diseases such as macular degeneration.