Agurto Carla, Yu Honggang, Murray Victor, Pattichis Marios S, Nemeth Sheila, Barriga Simon, Soliz Peter
Department of Electrical and Computer Engineering, University of New Mexico, Albuquerque, NM, USA; VisionQuest Biomedical LLC, Albuquerque, NM, USA.
VisionQuest Biomedical LLC, Albuquerque, NM, USA.
Comput Med Imaging Graph. 2015 Jul;43:137-49. doi: 10.1016/j.compmedimag.2015.01.001. Epub 2015 Jan 20.
This paper presents a multiscale method to detect neovascularization in the optic disc (NVD) using fundus images. Our method is applied to a manually selected region of interest (ROI) containing the optic disc. All the vessels in the ROI are segmented by adaptively combining contrast enhancement methods with a vessel segmentation technique. Textural features extracted using multiscale amplitude-modulation frequency-modulation, morphological granulometry, and fractal dimension are used. A linear SVM is used to perform the classification, which is tested by means of 10-fold cross-validation. The performance is evaluated using 300 images achieving an AUC of 0.93 with maximum accuracy of 88%.
本文提出了一种利用眼底图像检测视盘新生血管(NVD)的多尺度方法。我们的方法应用于手动选择的包含视盘的感兴趣区域(ROI)。通过将对比度增强方法与血管分割技术自适应地结合,对ROI中的所有血管进行分割。使用多尺度调幅调频、形态粒度分析和分形维数提取纹理特征。使用线性支持向量机进行分类,并通过10折交叉验证进行测试。使用300张图像评估性能,AUC为0.93,最大准确率为88%。