天天看点

【医学+深度论文:F32】2017 SPIE Automated detection of nerve fiber layer defects on retinal fundus32

32

2017 SPIE

Automated detection of nerve fiber layer defects on retinal fundus images using fully convolutional network for early diagnosis of glaucoma

Method : NFLD 分割

Dataset: Tajimi

Architecture: FCN-8s

Results: 最高 SE 98% , FP (5.42)

automated scheme for detection of a retinal nerve fiber layer defect(NFLD)

Method

Previous study

  • multi-step detection : Gabor filtering , clustering and adaptive thresholding
  • Problem :FP 多,method included too many rules

Deep convolutional neural network with fully convolutional layers

  • end-to-end : DCNN ( FCN-8s)
  • image
    • Original color images ofabnormal cases,
    • (b) original color images of both normal and abnormal cases,
    • © ellipse-based polar transformed colorimages,
    • (d) transformed G plane images,
    • (e) transformed Gabor filtered color images,
    • (f) transformed color halved images,
    • (g) transformed color halved images with different data augmentation.

      rotation and intensity transformation for all

  • add a softmax layer
【医学+深度论文:F32】2017 SPIE Automated detection of nerve fiber layer defects on retinal fundus32
【医学+深度论文:F32】2017 SPIE Automated detection of nerve fiber layer defects on retinal fundus32

Result

  • 使用 normal and abnormal image 减少了 FP 但是也降低了 SE
  • 虽然 绿色通道 对比度更高,但是用RGB图具有更高的灵敏度
  • DCNN 通用性比 previous study 更好
  • FROC ,最高SE 98% , FP (5.42)

Discussion

  • 研究了 不同输入图像 ,结果的不同,主要是从输入数据这块 做了一些对比实验(个人感觉含金量不高)
  • 研究了 FCN 在 眼底图中自动检测 NFLDs的应用

继续阅读