2sRanking-CNN: A 2-stage ranking-CNN for diagnosis of glaucoma from fundus images using CAM-extracted ROI as an intermediate input

  • 2018-05-15 12:27:00
  • Tae Joon Jun, Dohyeun Kim, Hoang Minh Nguyen, Daeyoung Kim, Youngsub Eom
  • 1

Abstract

Glaucoma is a disease in which the optic nerve is chronically damaged by theelevation of the intra-ocular pressure, resulting in visual field defect.Therefore, it is important to monitor and treat suspected patients before theyare confirmed with glaucoma. In this paper, we propose a 2-stage ranking-CNNthat classifies fundus images as normal, suspicious, and glaucoma. Furthermore,we propose a method of using the class activation map as a mask filter andcombining it with the original fundus image as an intermediate input. Ourresults have improved the average accuracy by about 10% over the existing3-class CNN and ranking-CNN, and especially improved the sensitivity ofsuspicious class by more than 20% over 3-class CNN. In addition, the extractedROI was also found to overlap with the diagnostic criteria of the physician.The method we propose is expected to be efficiently applied to any medical datawhere there is a suspicious condition between normal and disease.

 

Introduction (beta)

None

 

Conclusion (beta)

None