arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2205.03702eess.IVcs.CVcs.CY
收藏

用于智能手机角膜地形图仪的圆锥角膜分类器

Keratoconus Classifier for Smartphone-based Corneal Topographer

Microsoft Research(微软研究院);Sankara Eye Hospital(桑卡拉眼科医院)

Siddhartha Gairola, Pallavi Joshi, Anand Balasubramaniam, Kaushik Murali, Nipun Kwatra, Mohit Jain

摘要:

圆锥角膜是一种导致角膜变形的严重眼病。它影响10-25岁的人,是该人口统计中失明的主要原因。角膜地形图是圆锥角膜诊断的金标准。这是一个非侵入性的过程,使用昂贵和庞大的医疗设备称为角膜地形图。这使得大量人口无法获得,特别是在全球南部。已经提出了低成本的基于智能手机的角膜地形图仪,例如SmartKC,以使圆锥角膜诊断变得容易。与医用级地形图仪类似,SmartKC输出曲率热图和定量指标,需要医生评估以进行圆锥角膜诊断。用于评估这些热图和定量值的自动方案可以在医生不可用的地区筛查圆锥角膜中发挥关键作用。在这项工作中,我们提出了一种双头卷积神经网络(CNN),用于在SmartKC生成的热图上对圆锥角膜进行分类。由于SmartKC是一种新设备,只有一个小的数据集(114个样本),我们开发了一个2阶段的迁移学习策略-使用从医疗级地形仪收集的历史数据和SmartKC数据的子集-来令人满意地训练我们的网络。这一点,结合我们的特定领域的数据增强,实现了91.3%的灵敏度和94.2%的特异性。

英文摘要:

Keratoconus is a severe eye disease that leads to deformation of the cornea. It impacts people aged 10-25 years and is the leading cause of blindness in that demography. Corneal topography is the gold standard for keratoconus diagnosis. It is a non-invasive process performed using expensive and bulky medical devices called corneal topographers. This makes it inaccessible to large populations, especially in the Global South. Low-cost smartphone-based corneal topographers, such as SmartKC, have been proposed to make keratoconus diagnosis accessible. Similar to medical-grade topographers, SmartKC outputs curvature heatmaps and quantitative metrics that need to be evaluated by doctors for keratoconus diagnosis. An automatic scheme for evaluation of these heatmaps and quantitative values can play a crucial role in screening keratoconus in areas where doctors are not available. In this work, we propose a dual-head convolutional neural network (CNN) for classifying keratoconus on the heatmaps generated by SmartKC. Since SmartKC is a new device and only had a small dataset (114 samples), we developed a 2-stage transfer learning strategy -- using historical data collected from a medical-grade topographer and a subset of SmartKC data -- to satisfactorily train our network. This, combined with our domain-specific data augmentations, achieved a sensitivity of 91.3% and a specificity of 94.2%.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑