Department of Ophthalmology University Hospital, LMU Munich, Munich, Germany.
SMILE Eyes Clinic, Linz, Austria.
Curr Eye Res. 2024 Mar;49(3):252-259. doi: 10.1080/02713683.2023.2282938. Epub 2023 Nov 30.
AI (artificial intelligence)-based methodologies have become established tools for researchers and physicians in the entire field of ophthalmology. However, the potential of AI to optimize the refractive outcome of keratorefractive surgery by means of machine learning (ML)-based nomograms has not been exhausted yet. In this study, we wanted to comprehensively compare state-of-the-art conventional nomograms for Small-Incision-Lenticule-Extraction (SMILE) with a novel ML-based nomogram regarding both their spherical and astigmatic predictability.
A total of 1,342 eyes were analyzed for creation of three different nomograms based on a linear model (LM), a generalized additive mixed model (GAMM) and an artificial-neuronal-network (ANN), respectively. A total of 16 patient- and treatment-related features were included. Each model was trained by 895 eyes and validated by the remaining 447 eyes. Predictability was assessed by the difference between attempted and achieved change in spherical equivalent (SE) and the difference between target induced astigmatism (TIA) and surgically induced astigmatism (SIA). The root mean squared error (RMSE) of each model was computed as a measure of overall model performance.
The RMSE of LM, GAMM and ANN were 0.355, 0.348 and 0.367 for the prediction of SE and 0.279, 0.278 and 0.290 for the astigmatic correction, respectively. By applying the created models, the theoretical yield of eyes within ±0.50 D of SE from target refraction improved from 82 to 83% (LM), 84% (GAMM) and 83% (ANN), respectively. Astigmatic outcomes showed an improvement of eyes within ±0.50 D from TIA from 90 to 93% (LM), 93% (GAMM) and 92% (ANN), respectively. Subjective manifest refraction was the single most influential covariate in all models.
Machine learning endorsed the validity of state-of-the-art linear and non-linear SMILE nomograms. However, improving the accuracy of subjective manifest refraction seems warranted for optimizing ±0.50 D SE predictability beyond an apparent methodological 90% limit.
基于人工智能（AI）的方法已经成为整个眼科领域研究人员和医生的既定工具。然而，通过机器学习（ML）为基于图的预测优化准分子激光角膜屈光手术的折射结果的潜力尚未得到充分发挥。在这项研究中，我们想全面比较最先进的基于常规图的小切口角膜透镜切除术（SMILE）与一种新型基于 ML 的图在球镜和散光预测方面的性能。
基于线性模型（LM）、广义加性混合模型（GAMM）和人工神经网络（ANN），分别对 1342 只眼进行分析，以创建三个不同的图。共纳入 16 个与患者和治疗相关的特征。每个模型均由 895 只眼进行训练，由剩余的 447 只眼进行验证。通过尝试与实际球镜等效（SE）变化之间的差异以及目标诱导散光（TIA）与手术诱导散光（SIA）之间的差异来评估预测性。每个模型的均方根误差（RMSE）作为整体模型性能的衡量标准。
LM、GAMM 和 ANN 对 SE 预测的 RMSE 分别为 0.355、0.348 和 0.367，对散光矫正的 RMSE 分别为 0.279、0.278 和 0.290。通过应用所创建的模型，目标屈光矫正 SE 在±0.50 D 范围内的眼睛的理论产量从 82%提高到 83%（LM）、84%（GAMM）和 83%（ANN）。散光结果显示，TIA 在±0.50 D 范围内的眼睛从 90%提高到 93%（LM）、93%（GAMM）和 92%（ANN）。主观显微微镜检查是所有模型中最具影响力的协变量。
机器学习支持最先进的线性和非线性 SMILE 图的有效性。然而，为了优化±0.50 D SE 预测的准确性，似乎有必要改进主观显微微镜检查的准确性，以超越明显的 90%方法学限制。
