基于MLP神经网络的圆锥角膜辅助诊断北大核心CSCD

圆锥角膜在病变过程中会导致角膜中央部位向前凸出,使角膜呈现出圆锥形,而且会导致高度不规则近视和散光,对视力造成不同程度损害。疾病一般发生于青少年时期,为了能及时治疗避免病变严重,筛查区分圆锥角膜具有十分重要的意义。而且临床上对于圆锥角膜诊断通常是采用角膜地形图的方法,可以得到角膜形态学的改变,但是有一定的误诊率。目前研究发现,角膜力学特性改变先于形态学,所以本文从角膜生物力学角度出发,提出一种基于多层感知机(multi-layer perceptron,MLP)神经网络区分圆锥角膜的模型。首先,利用可视化生物力学分析仪(corneal visualization scheimpflug technology,Corvis-ST)测得角膜的生物力学视频,进行处理计算得到角膜生物力学参数作为数据集,其中包含正常角膜和圆锥角膜2种类别;然后,针对角膜生物力学参数数据集构建MLP神经网络模型,将70%数据集作为训练集,30%数据集作为测试集。在数据集上训练及测试的结果表明,该模型区分圆锥角膜的准确率为97.6%。

Keratoconus model for auxiliary diagnosis based on MLP neural network

Keratoconus causes the central cornea to bulge forward during the dise ase process,giving the cornea a conical shape,and leading to highly irregular myopia and astigmatism,causing damage of vision with different degrees.The disea se generally occurs in the adolescent period,in order to timely treat and to av oid serious lesions, it is of great signif icance to screen and distinguish keratoconus.In addition,clinical diagnosis of keratoconus is usually detected by co rneal topography,which can obtain morphological changes of the cornea,but ther e is a certain misdiagnosis rate.At present,it has been found that the change of mechanical properties of cornea is prior to morphology.Therefore,from the p erspective of corneal biomechanics,this paper proposed a model to distinguish k eratoconus based on multi-layer perceptron (MLP) neural network.Firstly,corneal visualization scheimpflug technology (Corvis -ST) was used to measure the biomechanical video of cornea,and corneal biomechanic al parameters were obtained as a data set,including normal cornea and keratocon us.Then,MLP neural network model was constructed for corneal biomechanical par ameter data sets,in which 70% data sets were used as training sets and 30% as t est sets.The results of training and testing on the datasets showed that the ac curacy of keratoconus differentiation was 97.6%.