基于机器学习预测快速角膜胶原交联手术效果  

Machine learning-based prediction of accelerated corneal collagen cross-linking surgery outcomes

作　　者：万奇[1] 陈蠡 魏然[1] 殷鸿波[1] 唐静[1] 邓应平[1] 马可[1] Wan Qi;Chen Li;Wei Ran;Yin Hongbo;Tang Jing;Deng Yingping;Ma Ke(Department of Ophthalmology,West China Hospital,Sichuan University,Chengdu 610041,China)

机构地区：[1]四川大学华西医院眼科,成都610041

出　　处：《中华实验眼科杂志》2025年第4期323-334,共12页Chinese Journal Of Experimental Ophthalmology

基　　金：四川省卫生健康委员会科技项目(24LCYJPT20)。

摘　　要：目的基于机器学习预测快速角膜胶原交联(A-CXL)手术效果,识别预后因素并构建模型预测术后疾病进展。方法采用单中心回顾性研究方法,收集2021年3—12月在四川大学华西医院眼科门诊接受A-CXL手术的圆锥角膜患者82例112眼。在术前和随访期间,采用裂隙灯显微镜检查眼前节情况,采用Pentacam检查角膜地形图,采用Corvis ST检查角膜生物力学参数。末次随访时最大角膜曲率(Kmax)较术前增加≥1 D定义为圆锥角膜进展。运用多种机器学习算法对角膜地形图、生物力学参数和角膜光密度值进行分析,识别预后因素并构建模型来预测术后疾病进展。结果在随访期间,15.1%(17/112)患眼出现A-CXL术后进展。进展组术前散光度和应力-应变指数(SSI)分别为(-5.41±2.72)D和1.41±0.78,分别大于非进展组的(-3.30±2.54)D和0.95±0.98,差异均有统计学意义(t=2.80、2.03,均P<0.05)。通过Cox回归分析发现,术前散光[风险比(HR)=1.20]、SSI(HR=1.10)和2~6 mm范围角膜前表面光密度(CDA6)(HR=2.10)是A-CXL术后进展的显著危险因素。通过多种机器学习模型的开发和对比验证发现,Logistic回归、多层感知器模型(MLP)和随机森林的曲线下面积(AUC)值均超过0.700。在F1分数评价方面,Logistic回归、MLP和RF的AUC值分别为0.870、0.880和0.880。可视化MLP的网络结构为含有24个神经元的单层神经网络,对A-CXL术后是否发生进展的预测准确率达到80%。结合散光、SSI和CDA6开发的临床列线图,可根据每个变量指定分数的总分数来预测术后0.5、1和2年的累积进展概率,根据最佳截断值,可以将圆锥角膜相应地分为高、中和低风险组。列线图时间依赖的受试者操作特征曲线显示,术后0.5、1和2年的AUC分别为0.734、0.685和0.935,预测概率均表现良好。结论术前散光、SSI和CDA6是圆锥角膜A-CXL术后进展的重要危险因素。MLP可以准确预测术后疾病进展,结合术前散光�Objective To use machine learning to predict the efficacy of accelerated corneal collagen cross-linking(A-CXL)surgery,identify prognostic factors,and construct models to predict postoperative disease progression.Methods A single-center retrospective study was conducted.A total of 82 keratoconus patients(112 eyes)who underwent A-CXL surgery at the West China Hospital of Sichuan University between March and December 2021 were enrolled.Preoperative and follow-up examinations included anterior segment evaluation by slit-lamp microscopy,corneal topography using Pentacam,and corneal biomechanical indices using Corvis ST.Disease progression was defined as an increase in maximum keratometry(Kmax)of≥1 D from the preoperative level at the last follow-up.Various machine learning algorithms were employed to analyze corneal topography,biomechanical parameters and corneal densitometry values to identify prognostic factors and construct models for predicting postoperative disease progression.This study adhered to the Declaration of Helsinki.The study protocol was approved by the Ethics Committee of West China Hospital,Sichuan University(No.2023496).Written informed consent was obtained from each subject.Results During follow-up,15.1%(17/112)of the eyes showed progression after A-CXL.The preoperative astigmatism and stress-strain index(SSI)in the progression group were(-5.41±2.72)D and 1.41±0.78,respectively,which were significantly higher than(-3.30±2.54)D and 0.95±0.98 in the non-progression group(t=2.80,2.03;both P<0.05).Cox regression analysis identified preoperative astigmatism(hazard ratio[HR]=1.20),SSI(HR=1.10),and anterior corneal densitometry of 2-6 mm(CDA6)(HR=2.10)as significant risk factors for post-A-CXL progression.Among various machine learning models developed and validated,the area under the curve(AUC)values for logistic regression,multilayer perceptron(MLP)model,and random forest(RF)exceeded 0.700.For F1-score,the AUC values for logistic regression,MLP,and RF were 0.870,0.880,and 0.880,respectively.The ne

关 键 词：圆锥角膜 交联手术 机器学习 预后预测 

分 类 号：R779.6[医药卫生—眼科]