Modelling subretinal injections to improve delivery of treatment for age-related eye diseases

老年性黄斑变性是导致60岁以上成年人严重视力丧失的主要原因之一。AMD是一种视网膜疾病，会导致视网膜中央区域(黄斑)的功能改变和减少，在最严重的情况下会导致中央视力丧失。由于解剖学和生理学的限制，将治疗药物输送到视网膜的这一区域是具有挑战性的。视网膜下注射是一种用于治疗视网膜疾病的各种创新眼科疗法的输送技术，例如基因疗法。视网膜下注射技术包括在RPE细胞和光感受器之间的间隙进行治疗。滤过泡的位置似乎受到几个因素的影响，包括相对于中心凹的位置，视网膜的厚度，视网膜血管的位置以及视网膜和RPE之间的粘连。靶向中心凹和扩大的气泡相对于注射点的不可预测的方向是该过程中持续存在的问题。计算模型是一种有用的工具，可以更好地了解在滤过泡的产生和传播中起关键作用的物理和解剖因素。使用经过验证的有限元模型可以是一种快速的方法，有助于改进视网膜下注射的技术，从而改善患者的预后并扩大眼科治疗的视野。这里的目的是开发一个眼睛的3D计算模型，其中包括解剖特征，如血管，以准确地再现视网膜下注射过程。为了建立准确描述的计算机模型，通过对视网膜条进行拉伸测试来确定视网膜的生物力学特性。为了比较视网膜不同区域的生物力学特性，从90只猪的眼睛中解剖了130个条状组织，每组10个样本。总共对13组不同的人进行了测试，以评估血管的影响、视网膜的各向异性特性以及象限之间的差异。研究结果表明，血管对视网膜的强度和硬度有显著影响。此外，在相同象限中提取的不同方向的条带显示出不同的抗张强度，这表明视网膜表面是各向异性的。利用ABAQUS有限元分析软件，建立了包含视网膜生物力学特性的二维有限元模型，模拟了视网膜与视网膜色素上皮(RPE)之间的粘连。通过参数研究评价了注射参数对滤过泡形状的影响，结果表明，眼压和注射速度显著影响滤过泡的形状，眼压的增加和注射速度的加快会导致较大的滤过泡高度。为了更好地模拟视网膜下滤过泡的生成，建立了一个结合了猪眼的几何和生物力学特性的3D有限元模型。3D模型的工作仍在进行中，重点是改进粘合层属性的定义，以生成与实际应用中观察到的高度和宽度相同的视网膜下气泡。这样，就可以使用视网膜下滤过泡的图像来验证模型，并执行参数研究来评估注射参数对滤过泡形状的影响。

Age-related macular degeneration is one of the main causes of severe vision loss in adults over 60. AMD is a retinal disease that causes an alteration and a reduction in the functionality of the central area of the retina (macula), which in the most severe cases causes the loss of central vision. Delivery of therapeutic drugs to this area of the retina is challenging due to anatomical and physiological limitations. Subretinal injection is a delivery technique for a variety of innovative ocular therapies used to treat retinal diseases, such as gene therapy. Subretinal injection technique consists in the delivery of treatment in the space between the RPE cells and the photoreceptors. The location of the bleb seems to be regulated by several factors including the location relative to the fovea, the thickness of the retina, the location of retinal blood vessels and the adhesion between retina and RPE. Targeting the fovea and the unpredictable direction of the enlarging bleb with respect to the injection point are persistent problems during this procedure. Computational model can be a useful tool to have a better understanding of the physical and anatomical factors that play a crucial role in bleb generation and propagation. Use of validated finite element modelling can be a fast method that can help to refine the technique of subretinal injections, thereby enhancing patient outcomes and expanding the horizons of ophthalmic therapies. Here the aim was to develop a 3D computational model of the eye that includes anatomical features, such as blood vessels, to accurately reproduce the subretinal injection procedure. In order to build accurately described computer models, the biomechanical properties of the retina were determined performing tensile test on retinal strips. A total of 130 strips, ten samples for each group, were dissected from 90 porcine eyes to compare the biomechanical properties in the different areas of the retina. A total of thirteen different groups have been tested, to evaluate the effects of the blood vessels, the anisotropic properties of the retina and the difference between the quadrants. The results obtained showed that blood vessels significantly impact the strength and stiffness of the retina. Additionally, strips extracted in the same quadrants, but with different orientations, showed different tensile strength suggesting that the retina is anisotropic across its surface. Using Abaqus CAE, a two-dimensional finite element model was created to mimic the adhesion between the retina and the retinal pigment epithelium (RPE), incorporating the biomechanical properties of the retina. A parametric study was performed to assess the impact of the injection parameters on bleb shape and the results obtained showed that IOP and the injection rate significantly influence bleb shape, where increased IOP and faster injection rate result in larger bleb heights. To better mimic the subretinal bleb generation, a 3D finite element model incorporating the geometry and biomechanical properties of a porcine eye has been developed. Work on the 3D model is still ongoing, focusing on improving the definition of the cohesive layer properties to generate a subretinal bleb with the same height and width as those observed in actual applications. In this way, it would be possible to validate the model using images of subretinal blebs and perform a parametric study to evaluate the effects of the injection parameters on the bleb shape.