Scholarlink

<p>The SYN-OCT dataset contains a total of 200,000 synthetic cross-sectional circumpapillary optical coherence tomography (OCT) images, comprising of 100,000 images from a generative image model for glaucoma eyes and 100,000 from a generative image model for healthy normal eyes. The generative image models were developed using real OCT imaging data acquired from study participants at the Singapore Eye Research Institute, and were trained with the respective real images from glaucoma or healthy eyes. Structural characteristics of the synthetic data were validated and found to be comparable with the real data using an automated segmentation approach. These measurements are provided together with the synthetic images. We envision this dataset to be useful for the development or validation of deep learning applications for glaucoma analysis and detection, and as a dataset to study the synthetic generation of medical images and their usability. This dataset is publicly available.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

SYN-OCT:A synthetic dataset of ocular optical coherence tomography images from healthy and glaucoma eyes

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月10日

查看全文
  • Damon Wong,
  • Ashish Jith Sreejith Kumar,
  • Rachel S. Chong,
  • Monisha E. Nongpiur,
  • Rahat Husain,
  • Tina Wong,
  • Shamira Perera,
  • Tin Aung,
  • Bingyao Tan,
  • Ching-Yu Cheng,
  • Eranga N. Vithana,
  • Jacqueline Chua,
  • Alina Popa Cherecheanu,
  • Leopold Schmetterer

摘要

The SYN-OCT dataset contains a total of 200,000 synthetic cross-sectional circumpapillary optical coherence tomography (OCT) images, comprising of 100,000 images from a generative image model for glaucoma eyes and 100,000 from a generative image model for healthy normal eyes. The generative image models were developed using real OCT imaging data acquired from study participants at the Singapore Eye Research Institute, and were trained with the respective real images from glaucoma or healthy eyes. Structural characteristics of the synthetic data were validated and found to be comparable with the real data using an automated segmentation approach. These measurements are provided together with the synthetic images. We envision this dataset to be useful for the development or validation of deep learning applications for glaucoma analysis and detection, and as a dataset to study the synthetic generation of medical images and their usability. This dataset is publicly available.