Sub-category Classifiers for Multiple-instance Learning and Its Application to Retinal Nerve Fiber Layer Visibility Classification

@inproceedings{Manivannan2016SubcategoryCF,
  title={Sub-category Classifiers for Multiple-instance Learning and Its Application to Retinal Nerve Fiber Layer Visibility Classification},
  author={Siyamalan Manivannan and Caroline O Cobb and Stephen Burgess and Emanuele Trucco},
  booktitle={MICCAI},
  year={2016}
}
We propose a novel multiple instance learning method to assess the visibility (visible/not visible) of the retinal nerve fiber layer (RNFL) in fundus camera images. Using only image-level labels, our approach learns to classify the images as well as to localize the RNFL visible regions. We transform the original feature space to a discriminative subspace, and learn a region-level classifier in that subspace. We propose a margin-based loss function to jointly learn this subspace and the region… CONTINUE READING

References

Publications referenced by this paper.
SHOWING 1-10 OF 14 REFERENCES

Computer-aided diagnosis from weak supervision: A benchmarking study

  • Comp. Med. Imag. and Graph.
  • 2015
VIEW 8 EXCERPTS
HIGHLY INFLUENTIAL

Empowering Multiple Instance Histopathology Cancer Diagnosis by Cell Graphs

Anonymous Authors
  • 2014
VIEW 6 EXCERPTS
HIGHLY INFLUENTIAL

Traffic Sign Recognition With Hinge Loss Trained Convolutional Neural Networks

  • IEEE Transactions on Intelligent Transportation Systems
  • 2014
VIEW 2 EXCERPTS